Browse the data
Every graded trial in the benchmark. Scan the heatmap, read the task catalog, or filter the trial table. Every cell opens the agent's full trajectory.
Task
Model
Wallet
882 trials| AI-text detection | Claude Opus 5.5 | $10 | 98.3 | $7.01 | $76.68 | 8h 53m | 199 |
| AI-text detection | Claude Opus 5.5 | $50 | 98.9 | $40.72 | $60.61 | 7h | 207 |
| AI-text detection | Claude Opus 5.5 | $200 | 99.2 | $80.08 | $69.16 | 7h 40m | 251 |
| AI-text detection | Claude Sonnet 5.5 | $10 | 96.1 | $8.13 | $28.31 | 4h 50m | 381 |
| AI-text detection | Claude Sonnet 5.5 | $50 | 96.1 | $46.29 | $21.98 | 6h 30m | 443 |
| AI-text detection | Claude Sonnet 5.5 | $200 | 94.9 | $81.50 | $22.31 | 4h 10m | 382 |
| AI-text detection | Claude Fable 5.1 | $10 | 71.9 | $9.12 | $24.90 | 3h 25m | 182 |
| AI-text detection | Claude Fable 5.1 | $50 | 83.6 | $25.94 | $64.04 | 3h 55m | 174 |
| AI-text detection | Claude Fable 5.1 | $200 | 83.2 | $123.89 | $98.16 | 5h 17m | 111 |
| AI-text detection | Claude Haiku 5.5 | $10 | 86.8 | $4.27 | $5.31 | 2h 26m | 267 |
| AI-text detection | Claude Haiku 5.5 | $50 | 93.8 | $26.76 | $7.03 | 5h 28m | 321 |
| AI-text detection | Claude Haiku 5.5 | $200 | 74.1 | $28.23 | $9.46 | 5h 50m | 388 |
| AI-text detection | GPT-6 Astra | $10 | 90.7 | $6.98 | $25.58 | 1h 11m | 133 |
| AI-text detection | GPT-6 Astra | $50 | 91.9 | $20.94 | $54.85 | 1h 48m | 284 |
| AI-text detection | GPT-6 Astra | $200 | 94.2 | $41.88 | $42.51 | 1h 59m | 210 |
| AI-text detection | GPT-6.1 Sol | $10 | 90.8 | $8.82 | $4.00 | 1h 2m | 134 |
| AI-text detection | GPT-6.1 Sol | $50 | 87.2 | $21.35 | $5.23 | 2h 30m | 182 |
| AI-text detection | GPT-6.1 Sol | $200 | 89.5 | $38.14 | $4.94 | 2h 7m | 158 |
| AI-text detection | GPT-6 Luna | $10 | 83.8 | $6.98 | $0.49 | 50m | 229 |
| AI-text detection | GPT-6 Luna | $50 | 84.0 | $5.76 | $0.42 | 49m | 246 |
| AI-text detection | GPT-6 Luna | $200 | 84.6 | $13.96 | $0.45 | 47m | 262 |
| AI-text detection | Gemini 3.8 Flash | $10 | 9.4 | $0.45 | $8.10 | 58m | 651 |
| AI-text detection | Gemini 3.8 Flash | $50 | 64.4 | $0.90 | $8.73 | 40m | 648 |
| AI-text detection | Gemini 3.8 Flash | $200 | 0.0 | $3.49 | $4.62 | 40m | 337 |
| AI-text detection | Grok 4.7 | $10 | 44.1 | $1.03 | $26.32 | 4h 1m | 100 |
| AI-text detection | Grok 4.7 | $50 | 83.0 | $2.36 | $25.70 | 2h 8m | 126 |
| AI-text detection | Grok 4.7 | $200 | 84.5 | $30.62 | $25.91 | 3h 58m | 177 |
| AI-text detection | Muse Spark 1.3 | $10 | 63.2 | $5.67 | $13.02 | 10h 14m | 331 |
| AI-text detection | Muse Spark 1.3 | $50 | 43.5 | $8.18 | $10.73 | 5h 29m | 228 |
| AI-text detection | Muse Spark 1.3 | $200 | 56.3 | $5.66 | $11.80 | 6h 39m | 240 |
| AI-text detection | Muse Wonder | $10 | 73.0 | $3.60 | $25.29 | 6h 9m | 284 |
| AI-text detection | Muse Wonder | $50 | 48.8 | $4.50 | $25.24 | 8h 13m | 282 |
| AI-text detection | Muse Wonder | $200 | 69.2 | $36.46 | $12.34 | 6h 24m | 213 |
| AI-text detection | Kimi K3 | $10 | 66.6 | $8.55 | $33.75 | 7h 13m | 370 |
| AI-text detection | Kimi K3 | $50 | 47.4 | $38.39 | $26.33 | 9h 26m | 357 |
| AI-text detection | Kimi K3 | $200 | 68.9 | $59.98 | $21.11 | 8h 41m | 321 |
| AI-text detection | Qwen3.8 Max | $10 | 76.5 | $8.51 | $20.36 | 10h 22m | 308 |
| AI-text detection | Qwen3.8 Max | $50 | 63.8 | $34.26 | $16.29 | 6h 34m | 441 |
| AI-text detection | Qwen3.8 Max | $200 | 54.1 | $15.42 | $8.34 | 4h 59m | 211 |
| AI-text detection | GLM 5.3 | $10 | 91.0 | $5.52 | $19.92 | 3h 32m | 290 |
| AI-text detection | GLM 5.3 | $50 | 55.2 | $21.85 | $10.33 | 9h 10m | 523 |
| AI-text detection | GLM 5.3 | $200 | 92.3 | $47.09 | $7.78 | 8h 6m | 422 |
| AR to diffusion conversion | Claude Opus 5.5 | $10 | 0.0 | $9.84 | $59.48 | 5h 13m | 333 |
| AR to diffusion conversion | Claude Opus 5.5 | $50 | 80.9 | $38.05 | $124.00 | 9h 29m | 337 |
| AR to diffusion conversion | Claude Opus 5.5 | $200 | 69.5 | $192.54 | $57.31 | 6h 25m | 328 |
| AR to diffusion conversion | Claude Sonnet 5.5 | $10 | 30.2 | $10.00 | $39.89 | 6h 44m | 593 |
| AR to diffusion conversion | Claude Sonnet 5.5 | $50 | 0.0 | $49.09 | $61.21 | 9h 18m | 1001 |
| AR to diffusion conversion | Claude Sonnet 5.5 | $200 | 47.6 | $169.27 | $50.21 | 10h 45m | 802 |
| AR to diffusion conversion | Claude Fable 5.1 | $10 | 0.0 | $9.90 | $73.42 | 6h 25m | 292 |
| AR to diffusion conversion | Claude Fable 5.1 | $50 | 76.7 | $26.88 | $54.54 | 4h 22m | 141 |
| AR to diffusion conversion | Claude Fable 5.1 | $200 | 27.9 | $167.52 | $122.44 | 5h 28m | 215 |
| AR to diffusion conversion | Claude Haiku 5.5 | $10 | 0.3 | $8.44 | $7.97 | 3h 45m | 318 |
| AR to diffusion conversion | Claude Haiku 5.5 | $50 | 12.3 | $28.89 | $15.98 | 6h | 344 |
| AR to diffusion conversion | Claude Haiku 5.5 | $200 | 39.2 | $153.56 | $10.69 | 4h 37m | 360 |
| AR to diffusion conversion | GPT-6 Astra | $10 | 4.8 | $9.89 | $67.43 | 2h 48m | 311 |
| AR to diffusion conversion | GPT-6 Astra | $50 | 2.1 | $49.98 | $128.18 | 4h 46m | 573 |
| AR to diffusion conversion | GPT-6 Astra | $200 | 0.0 | $199.59 | $205.65 | 9h 14m | 845 |
| AR to diffusion conversion | GPT-6.1 Sol | $10 | 1.7 | $9.83 | $7.66 | 3h 25m | 237 |
| AR to diffusion conversion | GPT-6.1 Sol | $50 | 55.3 | $48.86 | $16.32 | 9h 50m | 441 |
| AR to diffusion conversion | GPT-6.1 Sol | $200 | 45.9 | $197.77 | $17.24 | 9h 13m | 459 |
| AR to diffusion conversion | GPT-6 Luna | $10 | 0.0 | $3.49 | $0.55 | 1h 4m | 302 |
| AR to diffusion conversion | GPT-6 Luna | $50 | 2.1 | $38.39 | $3.25 | 10h 47m | 1774 |
| AR to diffusion conversion | GPT-6 Luna | $200 | 0.5 | $7.57 | $0.76 | 2h 10m | 359 |
| AR to diffusion conversion | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $28.96 | 12h | 1724 |
| AR to diffusion conversion | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $26.86 | 2h 26m | 1478 |
| AR to diffusion conversion | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $6.78 | 12h | 410 |
| AR to diffusion conversion | Grok 4.7 | $10 | 0.0 | $9.74 | $67.10 | 6h 9m | 163 |
| AR to diffusion conversion | Grok 4.7 | $50 | 0.0 | $34.90 | $60.37 | 5h 24m | 57 |
| AR to diffusion conversion | Grok 4.7 | $200 | 0.0 | $38.39 | $63.18 | 4h 14m | 283 |
| AR to diffusion conversion | Muse Spark 1.3 | $10 | 0.2 | $8.93 | $29.53 | 10h 52m | 600 |
| AR to diffusion conversion | Muse Spark 1.3 | $50 | 0.0 | $49.92 | $28.63 | 11h 55m | 606 |
| AR to diffusion conversion | Muse Spark 1.3 | $200 | 0.0 | $34.52 | $13.54 | 9h 5m | 324 |
| AR to diffusion conversion | Muse Wonder | $10 | 0.8 | $9.38 | $33.75 | 7h 1m | 393 |
| AR to diffusion conversion | Muse Wonder | $50 | 1.0 | $17.82 | $51.68 | 9h 59m | 547 |
| AR to diffusion conversion | Muse Wonder | $200 | 2.8 | $41.88 | $35.04 | 9h 23m | 369 |
| AR to diffusion conversion | Kimi K3 | $10 | 0.0 | $9.83 | $34.98 | 4h 45m | 311 |
| AR to diffusion conversion | Kimi K3 | $50 | 0.0 | $34.56 | $54.53 | 9h 2m | 461 |
| AR to diffusion conversion | Kimi K3 | $200 | 9.3 | $199.99 | $41.40 | 7h 56m | 406 |
| AR to diffusion conversion | Qwen3.8 Max | $10 | 0.0 | $9.95 | $1.28 | 6h 16m | 342 |
| AR to diffusion conversion | Qwen3.8 Max | $50 | 0.1 | $49.50 | $11.36 | 5h 55m | 235 |
| AR to diffusion conversion | Qwen3.8 Max | $200 | 0.0 | $76.66 | $4.43 | 12h | 385 |
| AR to diffusion conversion | GLM 5.3 | $10 | 0.0 | $8.63 | $5.34 | 2h 49m | 205 |
| AR to diffusion conversion | GLM 5.3 | $50 | 0.0 | $45.00 | $20.38 | 10h 6m | 795 |
| AR to diffusion conversion | GLM 5.3 | $200 | 0.0 | $65.69 | $16.27 | 12h 1m | 758 |
| Data-efficient pretraining | Claude Opus 5.5 | $10 | 65.1 | $9.60 | $26.53 | 3h 17m | 198 |
| Data-efficient pretraining | Claude Opus 5.5 | $50 | 83.1 | $41.88 | $42.77 | 6h 26m | 194 |
| Data-efficient pretraining | Claude Opus 5.5 | $200 | 99.8 | $129.13 | $25.49 | 5h 2m | 233 |
| Data-efficient pretraining | Claude Sonnet 5.5 | $10 | 47.8 | $7.53 | $28.87 | 6h 59m | 345 |
| Data-efficient pretraining | Claude Sonnet 5.5 | $50 | 71.1 | $34.90 | $29.08 | 9h 45m | 351 |
| Data-efficient pretraining | Claude Sonnet 5.5 | $200 | 80.5 | $108.19 | $25.94 | 10h 31m | 332 |
| Data-efficient pretraining | Claude Fable 5.1 | $10 | 26.9 | $5.59 | $48.86 | 3h 25m | 122 |
| Data-efficient pretraining | Claude Fable 5.1 | $50 | 40.9 | $24.43 | $102.30 | 5h 53m | 154 |
| Data-efficient pretraining | Claude Fable 5.1 | $200 | 80.6 | $34.90 | $91.52 | 5h 11m | 129 |
| Data-efficient pretraining | Claude Haiku 5.5 | $10 | 49.5 | $8.73 | $4.72 | 2h 30m | 152 |
| Data-efficient pretraining | Claude Haiku 5.5 | $50 | 67.2 | $26.18 | $12.76 | 6h 26m | 192 |
| Data-efficient pretraining | Claude Haiku 5.5 | $200 | 49.2 | $55.17 | $10.70 | 6h 13m | 254 |
| Data-efficient pretraining | GPT-6 Astra | $10 | 59.5 | $8.73 | $42.48 | 2h 11m | 281 |
| Data-efficient pretraining | GPT-6 Astra | $50 | 77.8 | $20.94 | $51.04 | 2h 17m | 235 |
| Data-efficient pretraining | GPT-6 Astra | $200 | 74.5 | $46.54 | $34.40 | 1h 59m | 201 |
| Data-efficient pretraining | GPT-6.1 Sol | $10 | 40.9 | $9.89 | $4.58 | 2h 33m | 134 |
| Data-efficient pretraining | GPT-6.1 Sol | $50 | 77.5 | $18.04 | $4.63 | 2h 34m | 150 |
| Data-efficient pretraining | GPT-6.1 Sol | $200 | 66.1 | $62.82 | $5.24 | 2h 49m | 134 |
| Data-efficient pretraining | GPT-6 Luna | $10 | 0.0 | $0.00 | $0.08 | 19m | 56 |
| Data-efficient pretraining | GPT-6 Luna | $50 | 0.0 | $0.00 | $0.09 | 20m | 58 |
| Data-efficient pretraining | GPT-6 Luna | $200 | 0.0 | $0.00 | $0.07 | 17m | 49 |
| Data-efficient pretraining | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $6.29 | 22m | 428 |
| Data-efficient pretraining | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $5.39 | 37m | 451 |
| Data-efficient pretraining | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $4.73 | 58m | 291 |
| Data-efficient pretraining | Grok 4.7 | $10 | 0.0 | $7.29 | $44.94 | 4h 1m | 273 |
| Data-efficient pretraining | Grok 4.7 | $50 | 0.0 | $5.40 | $39.73 | 8h 46m | 230 |
| Data-efficient pretraining | Grok 4.7 | $200 | 16.3 | $27.92 | $32.05 | 5h 40m | 117 |
| Data-efficient pretraining | Muse Spark 1.3 | $10 | 0.0 | $8.82 | $8.85 | 8h 7m | 214 |
| Data-efficient pretraining | Muse Spark 1.3 | $50 | 0.0 | $19.96 | $6.78 | 3h 50m | 196 |
| Data-efficient pretraining | Muse Spark 1.3 | $200 | 0.0 | $37.66 | $16.75 | 4h 52m | 318 |
| Data-efficient pretraining | Muse Wonder | $10 | 33.8 | $9.90 | $11.58 | 6h 2m | 187 |
| Data-efficient pretraining | Muse Wonder | $50 | 18.9 | $13.96 | $6.67 | 3h 41m | 130 |
| Data-efficient pretraining | Muse Wonder | $200 | 5.5 | $31.41 | $9.70 | 4h 22m | 165 |
| Data-efficient pretraining | Kimi K3 | $10 | 0.0 | $9.89 | $9.78 | 3h 29m | 158 |
| Data-efficient pretraining | Kimi K3 | $50 | 13.7 | $38.39 | $36.30 | 10h 17m | 420 |
| Data-efficient pretraining | Kimi K3 | $200 | 36.3 | $63.16 | $53.13 | 9h 29m | 383 |
| Data-efficient pretraining | Qwen3.8 Max | $10 | 34.0 | $8.63 | $8.52 | 4h 17m | 199 |
| Data-efficient pretraining | Qwen3.8 Max | $50 | 0.0 | $38.39 | $13.05 | 6h 43m | 277 |
| Data-efficient pretraining | Qwen3.8 Max | $200 | 49.5 | $40.47 | $13.78 | 7h 37m | 196 |
| Data-efficient pretraining | GLM 5.3 | $10 | 41.0 | $9.34 | $21.02 | 4h 50m | 259 |
| Data-efficient pretraining | GLM 5.3 | $50 | 37.8 | $16.87 | $17.44 | 4h 50m | 242 |
| Data-efficient pretraining | GLM 5.3 | $200 | 28.1 | $41.88 | $25.52 | 3h 52m | 374 |
| Figure to TikZ | Claude Opus 5.5 | $10 | 56.4 | $8.15 | $17.60 | 2h 31m | 188 |
| Figure to TikZ | Claude Opus 5.5 | $50 | 57.0 | $40.52 | $22.34 | 9h 12m | 327 |
| Figure to TikZ | Claude Opus 5.5 | $200 | 65.1 | $136.11 | $38.15 | 5h 5m | 239 |
| Figure to TikZ | Claude Sonnet 5.5 | $10 | 54.7 | $8.76 | $21.51 | 3h 41m | 455 |
| Figure to TikZ | Claude Sonnet 5.5 | $50 | 57.7 | $31.38 | $24.06 | 6h 58m | 549 |
| Figure to TikZ | Claude Sonnet 5.5 | $200 | 58.7 | $196.61 | $36.62 | 9h 12m | 482 |
| Figure to TikZ | Claude Fable 5.1 | $10 | 54.5 | $8.67 | $80.38 | 4h 46m | 119 |
| Figure to TikZ | Claude Fable 5.1 | $50 | 54.0 | $29.15 | $107.17 | 6h 32m | 145 |
| Figure to TikZ | Claude Fable 5.1 | $200 | 56.5 | $95.98 | $161.49 | 7h 53m | 183 |
| Figure to TikZ | Claude Haiku 5.5 | $10 | 42.4 | $7.81 | $7.01 | 3h 26m | 335 |
| Figure to TikZ | Claude Haiku 5.5 | $50 | 53.3 | $27.92 | $7.89 | 6h 34m | 382 |
| Figure to TikZ | Claude Haiku 5.5 | $200 | 52.2 | $41.88 | $10.96 | 5h 41m | 324 |
| Figure to TikZ | GPT-6 Astra | $10 | 52.9 | $10.00 | $182.35 | 8h 2m | 963 |
| Figure to TikZ | GPT-6 Astra | $50 | 57.0 | $49.74 | $248.86 | 11h | 1265 |
| Figure to TikZ | GPT-6 Astra | $200 | 59.1 | $160.54 | $109.27 | 4h 23m | 596 |
| Figure to TikZ | GPT-6.1 Sol | $10 | 50.5 | $9.99 | $6.72 | 3h 43m | 228 |
| Figure to TikZ | GPT-6.1 Sol | $50 | 51.0 | $39.56 | $9.59 | 8h 35m | 347 |
| Figure to TikZ | GPT-6.1 Sol | $200 | 60.4 | $195.44 | $10.79 | 6h 19m | 317 |
| Figure to TikZ | GPT-6 Luna | $10 | 41.4 | $5.76 | $0.86 | 1h 29m | 478 |
| Figure to TikZ | GPT-6 Luna | $50 | 45.7 | $34.90 | $0.94 | 4h 33m | 678 |
| Figure to TikZ | GPT-6 Luna | $200 | 32.9 | $34.52 | $1.34 | 6h 34m | 1018 |
| Figure to TikZ | Gemini 3.8 Flash | $10 | 30.6 | $2.48 | $9.02 | 1h 50m | 601 |
| Figure to TikZ | Gemini 3.8 Flash | $50 | 27.8 | $1.49 | $38.28 | 2h 27m | 2681 |
| Figure to TikZ | Gemini 3.8 Flash | $200 | 34.9 | $6.98 | $8.05 | 59m | 704 |
| Figure to TikZ | Grok 4.7 | $10 | 36.0 | $2.70 | $17.69 | 1h 57m | 114 |
| Figure to TikZ | Grok 4.7 | $50 | 35.5 | $11.88 | $24.44 | 7h 5m | 225 |
| Figure to TikZ | Grok 4.7 | $200 | 7.8 | $18.73 | $21.31 | 2h 58m | 31 |
| Figure to TikZ | Muse Spark 1.3 | $10 | 49.3 | $9.27 | $7.47 | 8h 10m | 220 |
| Figure to TikZ | Muse Spark 1.3 | $50 | 43.6 | $28.77 | $12.31 | 9h 2m | 272 |
| Figure to TikZ | Muse Spark 1.3 | $200 | 47.8 | $40.15 | $14.82 | 9h 11m | 314 |
| Figure to TikZ | Muse Wonder | $10 | 41.4 | $8.91 | $10.00 | 4h 40m | 175 |
| Figure to TikZ | Muse Wonder | $50 | 44.4 | $17.82 | $6.97 | 6h 58m | 149 |
| Figure to TikZ | Muse Wonder | $200 | 43.0 | $31.65 | $15.27 | 8h 14m | 235 |
| Figure to TikZ | Kimi K3 | $10 | 48.4 | $9.90 | $36.58 | 8h 21m | 407 |
| Figure to TikZ | Kimi K3 | $50 | 48.3 | $31.41 | $19.14 | 7h 56m | 257 |
| Figure to TikZ | Kimi K3 | $200 | 41.4 | $62.82 | $50.67 | 9h 41m | 419 |
| Figure to TikZ | Qwen3.8 Max | $10 | 44.4 | $7.80 | $11.91 | 5h 58m | 234 |
| Figure to TikZ | Qwen3.8 Max | $50 | 46.8 | $48.86 | $6.74 | 3h 56m | 158 |
| Figure to TikZ | Qwen3.8 Max | $200 | 56.3 | $115.06 | $16.77 | 9h | 404 |
| Figure to TikZ | GLM 5.3 | $10 | 39.5 | $9.92 | $5.63 | 3h 41m | 271 |
| Figure to TikZ | GLM 5.3 | $50 | 40.1 | $43.53 | $37.70 | 5h 31m | 562 |
| Figure to TikZ | GLM 5.3 | $200 | 49.0 | $48.91 | $29.80 | 8h 29m | 456 |
| Inverse dynamics model | Claude Opus 5.5 | $10 | 72.6 | $5.63 | $88.28 | 7h 29m | 330 |
| Inverse dynamics model | Claude Opus 5.5 | $50 | 66.9 | $25.30 | $62.88 | 7h 9m | 407 |
| Inverse dynamics model | Claude Opus 5.5 | $200 | 78.5 | $195.44 | $62.56 | 6h 55m | 265 |
| Inverse dynamics model | Claude Sonnet 5.5 | $10 | 73.3 | $7.86 | $54.78 | 8h 5m | 1051 |
| Inverse dynamics model | Claude Sonnet 5.5 | $50 | 69.0 | $10.07 | $120.89 | 9h 7m | 1783 |
| Inverse dynamics model | Claude Sonnet 5.5 | $200 | 79.1 | $180.94 | $247.24 | 7h 28m | 3484 |
| Inverse dynamics model | Claude Fable 5.1 | $10 | 56.8 | $2.70 | $97.90 | 4h 28m | 174 |
| Inverse dynamics model | Claude Fable 5.1 | $50 | 64.4 | $3.15 | $51.92 | 3h 44m | 178 |
| Inverse dynamics model | Claude Fable 5.1 | $200 | 63.2 | $34.90 | $22.56 | 1h 38m | 115 |
| Inverse dynamics model | Claude Haiku 5.5 | $10 | 32.2 | $7.86 | $4.14 | 2h 31m | 219 |
| Inverse dynamics model | Claude Haiku 5.5 | $50 | 43.2 | $16.58 | $11.18 | 4h 45m | 247 |
| Inverse dynamics model | Claude Haiku 5.5 | $200 | 37.4 | $13.96 | $6.94 | 3h 4m | 283 |
| Inverse dynamics model | GPT-6 Astra | $10 | 44.9 | $4.60 | $29.00 | 56m | 158 |
| Inverse dynamics model | GPT-6 Astra | $50 | 55.9 | $13.96 | $42.95 | 1h 9m | 205 |
| Inverse dynamics model | GPT-6 Astra | $200 | 67.2 | $20.14 | $34.22 | 1h 19m | 178 |
| Inverse dynamics model | GPT-6.1 Sol | $10 | 45.8 | $2.97 | $4.07 | 1h 16m | 115 |
| Inverse dynamics model | GPT-6.1 Sol | $50 | 53.7 | $17.45 | $4.84 | 1h 32m | 191 |
| Inverse dynamics model | GPT-6.1 Sol | $200 | 39.9 | $13.96 | $5.16 | 1h 49m | 182 |
| Inverse dynamics model | GPT-6 Luna | $10 | 36.5 | $0.00 | $0.39 | 1h 7m | 208 |
| Inverse dynamics model | GPT-6 Luna | $50 | 30.2 | $6.98 | $0.35 | 52m | 185 |
| Inverse dynamics model | GPT-6 Luna | $200 | 16.3 | $27.92 | $0.87 | 2h 6m | 533 |
| Inverse dynamics model | Gemini 3.8 Flash | $10 | 34.8 | $0.01 | $6.59 | 37m | 443 |
| Inverse dynamics model | Gemini 3.8 Flash | $50 | 31.4 | $5.20 | $14.73 | 1h 29m | 976 |
| Inverse dynamics model | Gemini 3.8 Flash | $200 | 16.1 | $3.72 | $23.90 | 2h 44m | 1542 |
| Inverse dynamics model | Grok 4.7 | $10 | 23.5 | $2.25 | $26.15 | 2h 14m | 189 |
| Inverse dynamics model | Grok 4.7 | $50 | 19.7 | $1.36 | $24.16 | 1h 31m | 195 |
| Inverse dynamics model | Grok 4.7 | $200 | 25.1 | $13.96 | $25.13 | 2h 24m | 116 |
| Inverse dynamics model | Muse Spark 1.3 | $10 | 19.3 | $2.70 | $34.75 | 6h 30m | 767 |
| Inverse dynamics model | Muse Spark 1.3 | $50 | 47.8 | $43.34 | $22.46 | 9h 34m | 529 |
| Inverse dynamics model | Muse Spark 1.3 | $200 | 46.0 | $35.18 | $20.95 | 11h 27m | 357 |
| Inverse dynamics model | Muse Wonder | $10 | 34.5 | $1.13 | $29.05 | 8h 24m | 359 |
| Inverse dynamics model | Muse Wonder | $50 | 48.1 | $33.82 | $31.02 | 10h 59m | 377 |
| Inverse dynamics model | Muse Wonder | $200 | 29.5 | $11.88 | $10.80 | 4h 46m | 176 |
| Inverse dynamics model | Kimi K3 | $10 | 34.7 | $2.88 | $67.27 | 10h 16m | 525 |
| Inverse dynamics model | Kimi K3 | $50 | 35.6 | $32.42 | $126.49 | 10h 15m | 878 |
| Inverse dynamics model | Kimi K3 | $200 | 31.6 | $39.92 | $75.94 | 11h 2m | 602 |
| Inverse dynamics model | Qwen3.8 Max | $10 | 36.0 | $2.70 | $8.90 | 3h 40m | 217 |
| Inverse dynamics model | Qwen3.8 Max | $50 | 67.5 | $31.21 | $17.26 | 10h 29m | 312 |
| Inverse dynamics model | Qwen3.8 Max | $200 | 57.3 | $172.18 | $12.30 | 5h 37m | 208 |
| Inverse dynamics model | GLM 5.3 | $10 | 34.4 | $5.40 | $42.18 | 7h 18m | 609 |
| Inverse dynamics model | GLM 5.3 | $50 | 35.0 | $30.42 | $17.93 | 9h 46m | 847 |
| Inverse dynamics model | GLM 5.3 | $200 | 34.1 | $15.51 | $59.78 | 7h 29m | 895 |
| Lean proof validity | Claude Opus 5.5 | $10 | 37.5 | $8.87 | $13.83 | 2h 52m | 192 |
| Lean proof validity | Claude Opus 5.5 | $50 | 45.3 | $35.49 | $31.22 | 6h 38m | 415 |
| Lean proof validity | Claude Opus 5.5 | $200 | 57.5 | $168.53 | $49.47 | 6h 2m | 282 |
| Lean proof validity | Claude Sonnet 5.5 | $10 | 41.8 | $8.41 | $30.63 | 9h 29m | 660 |
| Lean proof validity | Claude Sonnet 5.5 | $50 | 31.9 | $38.85 | $30.28 | 7h 28m | 522 |
| Lean proof validity | Claude Sonnet 5.5 | $200 | 57.9 | $167.54 | $30.62 | 6h 19m | 544 |
| Lean proof validity | Claude Fable 5.1 | $10 | 28.6 | $8.29 | $58.83 | 4h 4m | 82 |
| Lean proof validity | Claude Fable 5.1 | $50 | 40.7 | $31.50 | $96.25 | 5h 43m | 136 |
| Lean proof validity | Claude Fable 5.1 | $200 | 43.2 | $62.82 | $55.78 | 5h 33m | 199 |
| Lean proof validity | Claude Haiku 5.5 | $10 | 26.6 | $8.49 | $5.32 | 2h 59m | 299 |
| Lean proof validity | Claude Haiku 5.5 | $50 | 26.6 | $29.64 | $6.75 | 5h 22m | 350 |
| Lean proof validity | Claude Haiku 5.5 | $200 | 41.5 | $36.26 | $12.80 | 7h 28m | 393 |
| Lean proof validity | GPT-6 Astra | $10 | 37.5 | $9.15 | $138.49 | 5h 38m | 620 |
| Lean proof validity | GPT-6 Astra | $50 | 29.8 | $48.34 | $147.49 | 5h 41m | 696 |
| Lean proof validity | GPT-6 Astra | $200 | 35.8 | $169.24 | $146.10 | 6h 31m | 698 |
| Lean proof validity | GPT-6.1 Sol | $10 | 36.3 | $9.89 | $7.00 | 3h 1m | 210 |
| Lean proof validity | GPT-6.1 Sol | $50 | 27.7 | $48.27 | $13.49 | 8h 22m | 355 |
| Lean proof validity | GPT-6.1 Sol | $200 | 33.1 | $72.91 | $14.25 | 6h 15m | 342 |
| Lean proof validity | GPT-6 Luna | $10 | 1.4 | $6.24 | $0.45 | 49m | 250 |
| Lean proof validity | GPT-6 Luna | $50 | 15.7 | $8.63 | $0.88 | 1h 22m | 500 |
| Lean proof validity | GPT-6 Luna | $200 | 5.0 | $6.98 | $0.53 | 1h 2m | 292 |
| Lean proof validity | Gemini 3.8 Flash | $10 | 0.0 | $0.68 | $6.65 | 44m | 450 |
| Lean proof validity | Gemini 3.8 Flash | $50 | 0.0 | $8.43 | $19.55 | 4h 17m | 1206 |
| Lean proof validity | Gemini 3.8 Flash | $200 | 2.5 | $2.97 | $8.20 | 1h 8m | 579 |
| Lean proof validity | Grok 4.7 | $10 | 6.0 | $1.35 | $23.17 | 3h 4m | 49 |
| Lean proof validity | Grok 4.7 | $50 | 0.0 | $2.25 | $18.17 | 1h 43m | 226 |
| Lean proof validity | Grok 4.7 | $200 | 12.1 | $8.60 | $21.26 | 2h 41m | 118 |
| Lean proof validity | Muse Spark 1.3 | $10 | 4.3 | $2.91 | $12.99 | 5h 17m | 308 |
| Lean proof validity | Muse Spark 1.3 | $50 | 20.7 | $16.38 | $15.76 | 9h 14m | 356 |
| Lean proof validity | Muse Spark 1.3 | $200 | 0.0 | $36.48 | $25.69 | 12h | 529 |
| Lean proof validity | Muse Wonder | $10 | 0.0 | $2.00 | $3.81 | 2h 29m | 93 |
| Lean proof validity | Muse Wonder | $50 | 7.0 | $3.60 | $10.44 | 5h 22m | 194 |
| Lean proof validity | Muse Wonder | $200 | 0.1 | $29.85 | $12.98 | 8h 34m | 209 |
| Lean proof validity | Kimi K3 | $10 | 17.6 | $9.30 | $52.70 | 8h 53m | 555 |
| Lean proof validity | Kimi K3 | $50 | 17.2 | $42.77 | $34.97 | 10h 26m | 472 |
| Lean proof validity | Kimi K3 | $200 | 20.6 | $44.08 | $49.05 | 10h 14m | 520 |
| Lean proof validity | Qwen3.8 Max | $10 | 18.0 | $9.66 | $13.23 | 7h 41m | 354 |
| Lean proof validity | Qwen3.8 Max | $50 | 0.0 | $43.28 | $16.27 | 7h 21m | 357 |
| Lean proof validity | Qwen3.8 Max | $200 | 0.0 | $82.55 | $19.01 | 9h | 362 |
| Lean proof validity | GLM 5.3 | $10 | 0.0 | $9.30 | $31.30 | 5h 49m | 426 |
| Lean proof validity | GLM 5.3 | $50 | 0.0 | $22.28 | $39.82 | 7h 45m | 601 |
| Lean proof validity | GLM 5.3 | $200 | 3.0 | $35.35 | $25.71 | 9h 35m | 387 |
| LLM post-training | Claude Opus 5.5 | $10 | 57.6 | $7.86 | $33.22 | 3h 31m | 173 |
| LLM post-training | Claude Opus 5.5 | $50 | 43.3 | $44.21 | $18.50 | 4h 7m | 203 |
| LLM post-training | Claude Opus 5.5 | $200 | 67.4 | $173.34 | $53.77 | 6h 37m | 206 |
| LLM post-training | Claude Sonnet 5.5 | $10 | 49.2 | $9.92 | $23.19 | 4h 27m | 321 |
| LLM post-training | Claude Sonnet 5.5 | $50 | 49.3 | $46.00 | $39.87 | 9h 45m | 485 |
| LLM post-training | Claude Sonnet 5.5 | $200 | 45.2 | $184.01 | $40.86 | 8h 19m | 885 |
| LLM post-training | Claude Fable 5.1 | $10 | 26.0 | $6.98 | $38.90 | 2h 58m | 166 |
| LLM post-training | Claude Fable 5.1 | $50 | 46.2 | $48.86 | $47.79 | 3h 30m | 146 |
| LLM post-training | Claude Fable 5.1 | $200 | 0.0 | $181.48 | $86.57 | 5h 13m | 191 |
| LLM post-training | Claude Haiku 5.5 | $10 | 40.9 | $9.02 | $5.67 | 3h 11m | 206 |
| LLM post-training | Claude Haiku 5.5 | $50 | 17.4 | $37.82 | $18.70 | 8h 25m | 352 |
| LLM post-training | Claude Haiku 5.5 | $200 | 61.3 | $99.58 | $11.63 | 6h 54m | 426 |
| LLM post-training | GPT-6 Astra | $10 | 51.6 | $9.31 | $51.95 | 2h 44m | 250 |
| LLM post-training | GPT-6 Astra | $50 | 48.3 | $49.61 | $90.92 | 7h 8m | 575 |
| LLM post-training | GPT-6 Astra | $200 | 67.2 | $188.46 | $142.14 | 6h 42m | 713 |
| LLM post-training | GPT-6.1 Sol | $10 | 11.5 | $9.70 | $5.54 | 3h 1m | 182 |
| LLM post-training | GPT-6.1 Sol | $50 | 65.7 | $49.72 | $6.83 | 9h 39m | 238 |
| LLM post-training | GPT-6.1 Sol | $200 | 36.6 | $177.28 | $10.52 | 5h 27m | 361 |
| LLM post-training | GPT-6 Luna | $10 | 0.0 | $9.91 | $0.77 | 2h 10m | 449 |
| LLM post-training | GPT-6 Luna | $50 | 0.0 | $43.18 | $2.64 | 9h 17m | 1549 |
| LLM post-training | GPT-6 Luna | $200 | 0.0 | $41.88 | $1.91 | 8h 49m | 1286 |
| LLM post-training | Gemini 3.8 Flash | $10 | 0.0 | $0.68 | $8.59 | 1h 29m | 619 |
| LLM post-training | Gemini 3.8 Flash | $50 | 0.0 | $10.47 | $49.16 | 4h 46m | 3206 |
| LLM post-training | Gemini 3.8 Flash | $200 | 0.0 | $12.49 | $13.19 | 3h 49m | 757 |
| LLM post-training | Grok 4.7 | $10 | 0.0 | $9.59 | $21.77 | 3h 23m | 136 |
| LLM post-training | Grok 4.7 | $50 | 16.1 | $7.42 | $18.88 | 7h 23m | 193 |
| LLM post-training | Grok 4.7 | $200 | 0.0 | $34.53 | $23.65 | 12h | 208 |
| LLM post-training | Muse Spark 1.3 | $10 | 35.3 | $6.62 | $14.18 | 11h 1m | 281 |
| LLM post-training | Muse Spark 1.3 | $50 | 0.0 | $35.64 | $12.38 | 10h 25m | 315 |
| LLM post-training | Muse Spark 1.3 | $200 | 0.0 | $80.55 | $26.49 | 12h | 491 |
| LLM post-training | Muse Wonder | $10 | 0.0 | $8.42 | $15.69 | 5h 29m | 224 |
| LLM post-training | Muse Wonder | $50 | 34.8 | $34.90 | $26.29 | 9h 52m | 330 |
| LLM post-training | Muse Wonder | $200 | 48.8 | $153.56 | $18.83 | 9h 32m | 275 |
| LLM post-training | Kimi K3 | $10 | 13.3 | $9.83 | $18.56 | 5h 55m | 186 |
| LLM post-training | Kimi K3 | $50 | 26.1 | $41.88 | $24.50 | 10h 19m | 379 |
| LLM post-training | Kimi K3 | $200 | 12.9 | $197.37 | $47.57 | 11h 26m | 491 |
| LLM post-training | Qwen3.8 Max | $10 | 0.0 | $9.90 | $1.39 | 8h 3m | 283 |
| LLM post-training | Qwen3.8 Max | $50 | 38.1 | $49.85 | $24.91 | 10h 23m | 287 |
| LLM post-training | Qwen3.8 Max | $200 | 0.0 | $125.64 | $23.26 | 9h 37m | 591 |
| LLM post-training | GLM 5.3 | $10 | 24.9 | $8.73 | $19.73 | 2h 43m | 213 |
| LLM post-training | GLM 5.3 | $50 | 41.2 | $42.67 | $7.56 | 10h 52m | 318 |
| LLM post-training | GLM 5.3 | $200 | 34.3 | $37.65 | $3.85 | 7h 5m | 337 |
| Math proof simplification | Claude Opus 5.5 | $10 | 33.9 | $7.64 | $37.25 | 4h 19m | 275 |
| Math proof simplification | Claude Opus 5.5 | $50 | 29.4 | $23.50 | $64.32 | 5h 22m | 352 |
| Math proof simplification | Claude Opus 5.5 | $200 | 40.5 | $29.29 | $47.67 | 5h 16m | 281 |
| Math proof simplification | Claude Sonnet 5.5 | $10 | 31.6 | $6.52 | $32.01 | 5h 46m | 602 |
| Math proof simplification | Claude Sonnet 5.5 | $50 | 37.5 | $38.50 | $46.18 | 8h 31m | 987 |
| Math proof simplification | Claude Sonnet 5.5 | $200 | 34.9 | $37.97 | $31.15 | 7h 59m | 532 |
| Math proof simplification | Claude Fable 5.1 | $10 | 21.4 | $2.67 | $73.71 | 3h 39m | 188 |
| Math proof simplification | Claude Fable 5.1 | $50 | 29.8 | $9.11 | $38.17 | 2h 47m | 279 |
| Math proof simplification | Claude Fable 5.1 | $200 | 25.5 | $13.96 | $54.72 | 3h 14m | 192 |
| Math proof simplification | Claude Haiku 5.5 | $10 | 20.1 | $7.47 | $6.89 | 4h 11m | 323 |
| Math proof simplification | Claude Haiku 5.5 | $50 | 18.0 | $17.45 | $7.82 | 4h 56m | 295 |
| Math proof simplification | Claude Haiku 5.5 | $200 | 27.4 | $51.98 | $9.96 | 5h 31m | 406 |
| Math proof simplification | GPT-6 Astra | $10 | 21.5 | $3.49 | $29.31 | 58m | 156 |
| Math proof simplification | GPT-6 Astra | $50 | 28.0 | $10.47 | $39.75 | 1h 17m | 195 |
| Math proof simplification | GPT-6 Astra | $200 | 24.3 | $13.96 | $37.35 | 1h 40m | 191 |
| Math proof simplification | GPT-6.1 Sol | $10 | 15.6 | $3.49 | $2.79 | 53m | 123 |
| Math proof simplification | GPT-6.1 Sol | $50 | 14.8 | $8.63 | $3.91 | 1h 22m | 122 |
| Math proof simplification | GPT-6.1 Sol | $200 | 8.5 | $13.96 | $4.51 | 1h 31m | 141 |
| Math proof simplification | GPT-6 Luna | $10 | 2.8 | $0.00 | $0.41 | 45m | 206 |
| Math proof simplification | GPT-6 Luna | $50 | 0.1 | $10.47 | $0.49 | 1h 4m | 259 |
| Math proof simplification | GPT-6 Luna | $200 | 0.0 | $0.00 | $0.18 | 23m | 90 |
| Math proof simplification | Gemini 3.8 Flash | $10 | 4.6 | $0.01 | $10.28 | 53m | 754 |
| Math proof simplification | Gemini 3.8 Flash | $50 | 2.6 | $0.00 | $14.31 | 1h 4m | 995 |
| Math proof simplification | Gemini 3.8 Flash | $200 | 2.1 | $0.00 | $13.49 | 1h 24m | 1007 |
| Math proof simplification | Grok 4.7 | $10 | 5.3 | $7.36 | $35.11 | 2h 20m | 230 |
| Math proof simplification | Grok 4.7 | $50 | 6.5 | $0.00 | $25.68 | 2h 1m | 106 |
| Math proof simplification | Grok 4.7 | $200 | 2.2 | $9.87 | $24.86 | 1h 50m | 156 |
| Math proof simplification | Muse Spark 1.3 | $10 | 14.3 | $0.00 | $9.20 | 3h 2m | 198 |
| Math proof simplification | Muse Spark 1.3 | $50 | 20.5 | $0.01 | $13.80 | 4h 26m | 310 |
| Math proof simplification | Muse Spark 1.3 | $200 | 19.2 | $14.62 | $13.43 | 2h 26m | 250 |
| Math proof simplification | Muse Wonder | $10 | 11.4 | $8.91 | $9.06 | 2h 54m | 140 |
| Math proof simplification | Muse Wonder | $50 | 6.7 | $34.90 | $27.35 | 6h 55m | 341 |
| Math proof simplification | Muse Wonder | $200 | 3.3 | $29.44 | $23.15 | 4h 46m | 281 |
| Math proof simplification | Kimi K3 | $10 | 16.3 | $3.35 | $64.39 | 9h 33m | 629 |
| Math proof simplification | Kimi K3 | $50 | 11.2 | $10.54 | $26.44 | 4h 27m | 361 |
| Math proof simplification | Kimi K3 | $200 | 15.0 | $23.39 | $31.49 | 6h 7m | 360 |
| Math proof simplification | Qwen3.8 Max | $10 | 23.0 | $2.21 | $7.69 | 4h 5m | 241 |
| Math proof simplification | Qwen3.8 Max | $50 | 15.2 | $3.15 | $10.12 | 4h 51m | 314 |
| Math proof simplification | Qwen3.8 Max | $200 | 0.0 | $11.51 | $14.05 | 6h 26m | 403 |
| Math proof simplification | GLM 5.3 | $10 | 11.7 | $1.43 | $5.75 | 3h 20m | 295 |
| Math proof simplification | GLM 5.3 | $50 | 0.7 | $30.70 | $44.26 | 7h 40m | 635 |
| Math proof simplification | GLM 5.3 | $200 | 0.0 | $17.26 | $35.66 | 6h 28m | 529 |
| Music transcription | Claude Opus 5.5 | $10 | 61.9 | $7.39 | $98.13 | 9h 2m | 315 |
| Music transcription | Claude Opus 5.5 | $50 | 61.0 | $43.29 | $52.64 | 8h 4m | 501 |
| Music transcription | Claude Opus 5.5 | $200 | 64.3 | $85.65 | $68.74 | 6h 40m | 311 |
| Music transcription | Claude Sonnet 5.5 | $10 | 73.8 | $9.56 | $33.86 | 9h 44m | 529 |
| Music transcription | Claude Sonnet 5.5 | $50 | 79.5 | $18.93 | $41.57 | 9h 47m | 439 |
| Music transcription | Claude Sonnet 5.5 | $200 | 74.5 | $36.00 | $46.67 | 10h 11m | 469 |
| Music transcription | Claude Fable 5.1 | $10 | 28.8 | $5.52 | $28.81 | 5h 7m | 246 |
| Music transcription | Claude Fable 5.1 | $50 | 47.2 | $25.89 | $173.06 | 8h 8m | 124 |
| Music transcription | Claude Fable 5.1 | $200 | 28.1 | $37.95 | $127.55 | 6h 31m | 194 |
| Music transcription | Claude Haiku 5.5 | $10 | 0.0 | $8.39 | $4.23 | 3h 13m | 212 |
| Music transcription | Claude Haiku 5.5 | $50 | 0.0 | $24.43 | $7.77 | 4h 35m | 327 |
| Music transcription | Claude Haiku 5.5 | $200 | 0.0 | $30.71 | $8.06 | 3h 39m | 360 |
| Music transcription | GPT-6 Astra | $10 | 49.4 | $7.04 | $58.25 | 2h 13m | 295 |
| Music transcription | GPT-6 Astra | $50 | 73.4 | $46.01 | $153.20 | 5h 43m | 670 |
| Music transcription | GPT-6 Astra | $200 | 76.7 | $198.68 | $184.25 | 9h 57m | 828 |
| Music transcription | GPT-6.1 Sol | $10 | 51.1 | $5.30 | $8.34 | 3h 27m | 218 |
| Music transcription | GPT-6.1 Sol | $50 | 35.1 | $13.96 | $7.63 | 3h 8m | 232 |
| Music transcription | GPT-6.1 Sol | $200 | 73.1 | $199.32 | $14.29 | 9h 3m | 373 |
| Music transcription | GPT-6 Luna | $10 | 0.0 | $4.08 | $0.66 | 1h 22m | 349 |
| Music transcription | GPT-6 Luna | $50 | 0.0 | $6.98 | $0.30 | 36m | 161 |
| Music transcription | GPT-6 Luna | $200 | 0.0 | $1.63 | $0.74 | 1h 24m | 450 |
| Music transcription | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $15.07 | 1h 27m | 1111 |
| Music transcription | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $22.29 | 1h 48m | 1462 |
| Music transcription | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $10.68 | 1h 33m | 807 |
| Music transcription | Grok 4.7 | $10 | 0.0 | $2.03 | $26.52 | 2h 38m | 155 |
| Music transcription | Grok 4.7 | $50 | 0.0 | $10.47 | $31.49 | 2h 19m | 87 |
| Music transcription | Grok 4.7 | $200 | 1.6 | $12.18 | $51.61 | 2h 58m | 112 |
| Music transcription | Muse Spark 1.3 | $10 | 0.0 | $2.49 | $7.39 | 7h 20m | 208 |
| Music transcription | Muse Spark 1.3 | $50 | 0.0 | $3.60 | $12.78 | 5h 2m | 258 |
| Music transcription | Muse Spark 1.3 | $200 | 0.0 | $66.11 | $19.23 | 11h 2m | 339 |
| Music transcription | Muse Wonder | $10 | 4.6 | $5.16 | $32.31 | 4h 17m | 364 |
| Music transcription | Muse Wonder | $50 | 0.0 | $10.80 | $30.55 | 7h 5m | 387 |
| Music transcription | Muse Wonder | $200 | 13.3 | $2.70 | $7.29 | 4h 14m | 173 |
| Music transcription | Kimi K3 | $10 | 2.6 | $9.45 | $54.17 | 10h 17m | 484 |
| Music transcription | Kimi K3 | $50 | 0.0 | $13.37 | $37.60 | 8h 1m | 412 |
| Music transcription | Kimi K3 | $200 | 9.5 | $33.08 | $38.72 | 10h 8m | 429 |
| Music transcription | Qwen3.8 Max | $10 | 8.2 | $6.58 | $15.22 | 7h 45m | 424 |
| Music transcription | Qwen3.8 Max | $50 | 0.0 | $27.00 | $25.66 | 9h 7m | 806 |
| Music transcription | Qwen3.8 Max | $200 | 0.0 | $72.00 | $16.39 | 9h 15m | 317 |
| Music transcription | GLM 5.3 | $10 | 0.0 | $3.60 | $11.33 | 6h 53m | 547 |
| Music transcription | GLM 5.3 | $50 | 0.0 | $14.96 | $41.06 | 8h 32m | 572 |
| Music transcription | GLM 5.3 | $200 | 0.0 | $10.40 | $5.17 | 4h 30m | 265 |
| OCR latent compression | Claude Opus 5.5 | $10 | 26.9 | $9.88 | $70.68 | 5h 57m | 308 |
| OCR latent compression | Claude Opus 5.5 | $50 | 54.4 | $40.14 | $121.44 | 10h | 372 |
| OCR latent compression | Claude Opus 5.5 | $200 | 82.6 | $137.32 | $97.24 | 9h 8m | 202 |
| OCR latent compression | Claude Sonnet 5.5 | $10 | 55.8 | $9.90 | $30.38 | 7h 24m | 274 |
| OCR latent compression | Claude Sonnet 5.5 | $50 | 67.5 | $42.52 | $61.94 | 11h 5m | 649 |
| OCR latent compression | Claude Sonnet 5.5 | $200 | 63.9 | $198.87 | $60.25 | 8h 7m | 616 |
| OCR latent compression | Claude Fable 5.1 | $10 | 56.1 | $9.66 | $162.10 | 6h 58m | 200 |
| OCR latent compression | Claude Fable 5.1 | $50 | 67.9 | $39.36 | $199.56 | 7h 38m | 150 |
| OCR latent compression | Claude Fable 5.1 | $200 | 67.8 | $76.49 | $183.84 | 7h 54m | 301 |
| OCR latent compression | Claude Haiku 5.5 | $10 | 8.6 | $9.02 | $6.16 | 3h 17m | 246 |
| OCR latent compression | Claude Haiku 5.5 | $50 | 13.8 | $35.78 | $14.01 | 9h 20m | 440 |
| OCR latent compression | Claude Haiku 5.5 | $200 | 30.1 | $173.78 | $11.14 | 6h 3m | 401 |
| OCR latent compression | GPT-6 Astra | $10 | 36.4 | $8.78 | $73.28 | 3h 15m | 371 |
| OCR latent compression | GPT-6 Astra | $50 | 57.5 | $45.37 | $133.09 | 5h 50m | 618 |
| OCR latent compression | GPT-6 Astra | $200 | 64.1 | $150.20 | $242.87 | 10h 7m | 1162 |
| OCR latent compression | GPT-6.1 Sol | $10 | 37.6 | $9.94 | $13.98 | 6h 29m | 419 |
| OCR latent compression | GPT-6.1 Sol | $50 | 43.7 | $34.90 | $11.90 | 7h 10m | 333 |
| OCR latent compression | GPT-6.1 Sol | $200 | 35.2 | $125.64 | $10.28 | 8h 2m | 352 |
| OCR latent compression | GPT-6 Luna | $10 | 2.1 | $8.63 | $1.37 | 2h 38m | 708 |
| OCR latent compression | GPT-6 Luna | $50 | 0.4 | $5.76 | $0.58 | 1h 14m | 347 |
| OCR latent compression | GPT-6 Luna | $200 | 0.7 | $27.92 | $0.47 | 1h 20m | 293 |
| OCR latent compression | Gemini 3.8 Flash | $10 | 2.2 | $0.00 | $39.76 | 3h 48m | 2881 |
| OCR latent compression | Gemini 3.8 Flash | $50 | 0.0 | $0.90 | $14.29 | 2h 3m | 1010 |
| OCR latent compression | Gemini 3.8 Flash | $200 | 0.8 | $7.18 | $21.09 | 3h 37m | 1598 |
| OCR latent compression | Grok 4.7 | $10 | 26.1 | $1.24 | $49.14 | 3h 36m | 173 |
| OCR latent compression | Grok 4.7 | $50 | 4.9 | $20.94 | $54.89 | 4h 26m | 141 |
| OCR latent compression | Grok 4.7 | $200 | 4.8 | $41.88 | $31.69 | 4h 9m | 188 |
| OCR latent compression | Muse Spark 1.3 | $10 | 0.0 | $5.40 | $24.06 | 11h 58m | 556 |
| OCR latent compression | Muse Spark 1.3 | $50 | 0.0 | $6.08 | $20.73 | 12h | 457 |
| OCR latent compression | Muse Spark 1.3 | $200 | 0.0 | $46.38 | $24.76 | 12h | 383 |
| OCR latent compression | Muse Wonder | $10 | 15.3 | $8.63 | $6.64 | 2h 41m | 131 |
| OCR latent compression | Muse Wonder | $50 | 3.3 | $5.40 | $20.63 | 9h 8m | 272 |
| OCR latent compression | Muse Wonder | $200 | 1.9 | $41.88 | $27.88 | 9h 7m | 296 |
| OCR latent compression | Kimi K3 | $10 | 7.1 | $9.98 | $80.00 | 10h 49m | 702 |
| OCR latent compression | Kimi K3 | $50 | 15.5 | $14.85 | $92.69 | 10h 32m | 574 |
| OCR latent compression | Kimi K3 | $200 | 0.0 | $72.68 | $84.37 | 11h 26m | 559 |
| OCR latent compression | Qwen3.8 Max | $10 | 6.6 | $4.42 | $15.38 | 7h 50m | 293 |
| OCR latent compression | Qwen3.8 Max | $50 | 4.6 | $28.80 | $17.64 | 7h 46m | 532 |
| OCR latent compression | Qwen3.8 Max | $200 | 5.3 | $41.88 | $37.53 | 12h | 336 |
| OCR latent compression | GLM 5.3 | $10 | 4.2 | $8.91 | $45.69 | 8h 7m | 604 |
| OCR latent compression | GLM 5.3 | $50 | 6.3 | $42.42 | $22.42 | 12h | 1092 |
| OCR latent compression | GLM 5.3 | $200 | 5.7 | $33.56 | $10.59 | 9h 39m | 521 |
| PII redaction | Claude Opus 5.5 | $10 | 73.1 | $8.22 | $19.24 | 3h 53m | 225 |
| PII redaction | Claude Opus 5.5 | $50 | 77.3 | $25.87 | $20.64 | 3h 13m | 251 |
| PII redaction | Claude Opus 5.5 | $200 | 77.9 | $175.11 | $20.80 | 3h 12m | 251 |
| PII redaction | Claude Sonnet 5.5 | $10 | 70.6 | $5.86 | $31.12 | 6h 59m | 523 |
| PII redaction | Claude Sonnet 5.5 | $50 | 72.2 | $36.43 | $32.40 | 7h 13m | 432 |
| PII redaction | Claude Sonnet 5.5 | $200 | 68.4 | $103.55 | $27.92 | 6h 24m | 678 |
| PII redaction | Claude Fable 5.1 | $10 | 74.8 | $7.47 | $44.32 | 3h 13m | 173 |
| PII redaction | Claude Fable 5.1 | $50 | 70.0 | $34.56 | $34.12 | 2h 19m | 138 |
| PII redaction | Claude Fable 5.1 | $200 | 71.3 | $115.52 | $28.18 | 2h 56m | 230 |
| PII redaction | Claude Haiku 5.5 | $10 | 64.3 | $4.61 | $6.37 | 3h 4m | 294 |
| PII redaction | Claude Haiku 5.5 | $50 | 58.9 | $12.40 | $6.78 | 2h 56m | 305 |
| PII redaction | Claude Haiku 5.5 | $200 | 69.4 | $22.59 | $8.50 | 4h 25m | 370 |
| PII redaction | GPT-6 Astra | $10 | 64.8 | $3.49 | $26.00 | 49m | 147 |
| PII redaction | GPT-6 Astra | $50 | 68.0 | $24.22 | $41.62 | 1h 26m | 202 |
| PII redaction | GPT-6 Astra | $200 | 59.0 | $10.95 | $18.04 | 41m | 92 |
| PII redaction | GPT-6.1 Sol | $10 | 60.5 | $5.76 | $4.27 | 1h 13m | 150 |
| PII redaction | GPT-6.1 Sol | $50 | 64.3 | $12.30 | $8.30 | 2h 37m | 230 |
| PII redaction | GPT-6.1 Sol | $200 | 55.7 | $14.39 | $5.41 | 1h 46m | 197 |
| PII redaction | GPT-6 Luna | $10 | 21.8 | $3.49 | $0.26 | 20m | 142 |
| PII redaction | GPT-6 Luna | $50 | 18.5 | $0.00 | $0.30 | 37m | 142 |
| PII redaction | GPT-6 Luna | $200 | 17.4 | $0.00 | $0.18 | 17m | 99 |
| PII redaction | Gemini 3.8 Flash | $10 | 34.8 | $0.12 | $17.90 | 1h 19m | 1239 |
| PII redaction | Gemini 3.8 Flash | $50 | 49.8 | $0.45 | $8.81 | 1h 8m | 614 |
| PII redaction | Gemini 3.8 Flash | $200 | 33.8 | $3.49 | $6.05 | 1h 1m | 434 |
| PII redaction | Grok 4.7 | $10 | 43.8 | $0.95 | $21.99 | 1h 53m | 93 |
| PII redaction | Grok 4.7 | $50 | 0.0 | $12.22 | $17.80 | 1h 20m | 112 |
| PII redaction | Grok 4.7 | $200 | 31.5 | $4.71 | $22.78 | 1h 48m | 155 |
| PII redaction | Muse Spark 1.3 | $10 | 0.0 | $4.05 | $28.13 | 7h 57m | 552 |
| PII redaction | Muse Spark 1.3 | $50 | 35.0 | $24.79 | $24.73 | 10h 53m | 491 |
| PII redaction | Muse Spark 1.3 | $200 | 41.1 | $11.51 | $6.94 | 1h 14m | 171 |
| PII redaction | Muse Wonder | $10 | 67.3 | $1.80 | $11.77 | 2h 55m | 189 |
| PII redaction | Muse Wonder | $50 | 39.3 | $2.80 | $13.28 | 3h 3m | 235 |
| PII redaction | Muse Wonder | $200 | 63.4 | $3.15 | $19.01 | 3h 40m | 289 |
| PII redaction | Kimi K3 | $10 | 36.3 | $8.31 | $83.55 | 11h 9m | 532 |
| PII redaction | Kimi K3 | $50 | 56.1 | $31.41 | $23.14 | 4h 20m | 317 |
| PII redaction | Kimi K3 | $200 | 65.5 | $24.34 | $36.94 | 5h 27m | 343 |
| PII redaction | Qwen3.8 Max | $10 | 52.6 | $4.63 | $9.52 | 4h 50m | 210 |
| PII redaction | Qwen3.8 Max | $50 | 57.9 | $17.20 | $11.84 | 5h 19m | 249 |
| PII redaction | Qwen3.8 Max | $200 | 52.5 | $23.02 | $39.09 | 5h 17m | 765 |
| PII redaction | GLM 5.3 | $10 | 51.0 | $7.43 | $4.46 | 4h 9m | 206 |
| PII redaction | GLM 5.3 | $50 | 39.6 | $5.94 | $23.65 | 3h 49m | 356 |
| PII redaction | GLM 5.3 | $200 | 64.6 | $16.35 | $23.70 | 4h 19m | 357 |
| Pokémon Random Battle | Claude Opus 5.5 | $10 | 0.0 | $2.91 | $40.02 | 3h 15m | 291 |
| Pokémon Random Battle | Claude Opus 5.5 | $50 | 7.4 | $21.33 | $56.34 | 4h 59m | 473 |
| Pokémon Random Battle | Claude Opus 5.5 | $200 | 2.7 | $111.20 | $68.27 | 4h 33m | 401 |
| Pokémon Random Battle | Claude Sonnet 5.5 | $10 | 0.0 | $3.32 | $38.54 | 8h 42m | 507 |
| Pokémon Random Battle | Claude Sonnet 5.5 | $50 | 11.9 | $25.37 | $35.14 | 6h 10m | 497 |
| Pokémon Random Battle | Claude Sonnet 5.5 | $200 | 0.0 | $32.82 | $25.50 | 4h 1m | 501 |
| Pokémon Random Battle | Claude Fable 5.1 | $10 | 0.1 | $1.30 | $35.45 | 3h 30m | 238 |
| Pokémon Random Battle | Claude Fable 5.1 | $50 | 1.0 | $13.47 | $40.68 | 2h 41m | 153 |
| Pokémon Random Battle | Claude Fable 5.1 | $200 | 1.5 | $14.29 | $69.93 | 2h 55m | 85 |
| Pokémon Random Battle | Claude Haiku 5.5 | $10 | 1.1 | $2.65 | $7.08 | 3h 32m | 291 |
| Pokémon Random Battle | Claude Haiku 5.5 | $50 | 0.0 | $4.68 | $5.08 | 2h 47m | 229 |
| Pokémon Random Battle | Claude Haiku 5.5 | $200 | 0.0 | $1.60 | $6.89 | 2h 38m | 278 |
| Pokémon Random Battle | GPT-6 Astra | $10 | 6.4 | $5.54 | $181.13 | 6h 36m | 824 |
| Pokémon Random Battle | GPT-6 Astra | $50 | 7.2 | $49.78 | $267.82 | 10h 54m | 1301 |
| Pokémon Random Battle | GPT-6 Astra | $200 | 3.1 | $55.47 | $156.09 | 6h 22m | 640 |
| Pokémon Random Battle | GPT-6.1 Sol | $10 | 0.0 | $0.77 | $8.21 | 3h 34m | 225 |
| Pokémon Random Battle | GPT-6.1 Sol | $50 | 1.8 | $3.45 | $23.08 | 10h 22m | 564 |
| Pokémon Random Battle | GPT-6.1 Sol | $200 | 0.0 | $169.36 | $19.03 | 9h 16m | 486 |
| Pokémon Random Battle | GPT-6 Luna | $10 | 0.0 | $0.00 | $0.33 | 37m | 170 |
| Pokémon Random Battle | GPT-6 Luna | $50 | 0.0 | $0.00 | $0.27 | 35m | 139 |
| Pokémon Random Battle | GPT-6 Luna | $200 | 0.0 | $0.00 | $0.34 | 31m | 178 |
| Pokémon Random Battle | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $8.39 | 38m | 586 |
| Pokémon Random Battle | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $8.56 | 44m | 598 |
| Pokémon Random Battle | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $6.83 | 52m | 474 |
| Pokémon Random Battle | Grok 4.7 | $10 | 0.0 | $0.00 | $20.05 | 1h 26m | 235 |
| Pokémon Random Battle | Grok 4.7 | $50 | 0.0 | $3.04 | $31.67 | 2h 29m | 193 |
| Pokémon Random Battle | Grok 4.7 | $200 | 0.0 | $0.00 | $28.79 | 2h 2m | 120 |
| Pokémon Random Battle | Muse Spark 1.3 | $10 | 0.0 | $2.06 | $15.02 | 4h 16m | 368 |
| Pokémon Random Battle | Muse Spark 1.3 | $50 | 0.0 | $2.13 | $18.62 | 2h 19m | 388 |
| Pokémon Random Battle | Muse Spark 1.3 | $200 | 0.0 | $7.82 | $25.62 | 6h 10m | 554 |
| Pokémon Random Battle | Muse Wonder | $10 | 0.0 | $0.32 | $15.80 | 5h 9m | 235 |
| Pokémon Random Battle | Muse Wonder | $50 | 0.0 | $0.36 | $16.13 | 2h 54m | 203 |
| Pokémon Random Battle | Muse Wonder | $200 | 0.0 | $0.83 | $11.00 | 2h 50m | 174 |
| Pokémon Random Battle | Kimi K3 | $10 | 0.0 | $1.97 | $63.71 | 8h 11m | 581 |
| Pokémon Random Battle | Kimi K3 | $50 | 0.0 | $17.84 | $55.46 | 6h 35m | 529 |
| Pokémon Random Battle | Kimi K3 | $200 | 0.0 | $21.23 | $90.55 | 11h 9m | 715 |
| Pokémon Random Battle | Qwen3.8 Max | $10 | 0.0 | $0.34 | $8.40 | 4h 2m | 273 |
| Pokémon Random Battle | Qwen3.8 Max | $50 | 0.0 | $0.49 | $18.77 | 8h 38m | 505 |
| Pokémon Random Battle | Qwen3.8 Max | $200 | 0.0 | $2.63 | $10.43 | 5h 7m | 297 |
| Pokémon Random Battle | GLM 5.3 | $10 | 0.0 | $3.49 | $11.89 | 4h 50m | 616 |
| Pokémon Random Battle | GLM 5.3 | $50 | 0.0 | $1.86 | $16.08 | 7h 40m | 756 |
| Pokémon Random Battle | GLM 5.3 | $200 | 0.0 | $0.49 | $9.39 | 4h 40m | 445 |
| Screenshot diffing | Claude Opus 5.5 | $10 | 26.8 | $8.56 | $31.00 | 3h 25m | 307 |
| Screenshot diffing | Claude Opus 5.5 | $50 | 34.4 | $41.69 | $57.48 | 7h 8m | 296 |
| Screenshot diffing | Claude Opus 5.5 | $200 | 38.0 | $172.35 | $69.50 | 6h 19m | 354 |
| Screenshot diffing | Claude Sonnet 5.5 | $10 | 29.7 | $6.82 | $23.28 | 4h 43m | 355 |
| Screenshot diffing | Claude Sonnet 5.5 | $50 | 33.5 | $43.14 | $20.08 | 4h 43m | 269 |
| Screenshot diffing | Claude Sonnet 5.5 | $200 | 35.2 | $121.33 | $33.97 | 5h 48m | 582 |
| Screenshot diffing | Claude Fable 5.1 | $10 | 31.1 | $9.80 | $56.00 | 3h 24m | 112 |
| Screenshot diffing | Claude Fable 5.1 | $50 | 23.2 | $29.38 | $26.27 | 3h 4m | 227 |
| Screenshot diffing | Claude Fable 5.1 | $200 | 46.4 | $150.36 | $117.00 | 5h 46m | 225 |
| Screenshot diffing | Claude Haiku 5.5 | $10 | 19.2 | $9.20 | $5.00 | 2h 50m | 239 |
| Screenshot diffing | Claude Haiku 5.5 | $50 | 29.6 | $34.64 | $11.50 | 7h 5m | 294 |
| Screenshot diffing | Claude Haiku 5.5 | $200 | 38.5 | $70.65 | $8.71 | 5h 21m | 360 |
| Screenshot diffing | GPT-6 Astra | $10 | 16.0 | $6.98 | $40.79 | 1h 23m | 230 |
| Screenshot diffing | GPT-6 Astra | $50 | 15.5 | $13.96 | $40.56 | 1h 12m | 213 |
| Screenshot diffing | GPT-6 Astra | $200 | 8.2 | $33.57 | $48.88 | 1h 56m | 230 |
| Screenshot diffing | GPT-6.1 Sol | $10 | 6.2 | $5.82 | $4.18 | 1h 38m | 132 |
| Screenshot diffing | GPT-6.1 Sol | $50 | 10.2 | $17.26 | $7.17 | 2h 27m | 229 |
| Screenshot diffing | GPT-6.1 Sol | $200 | 10.5 | $27.98 | $8.10 | 3h 2m | 244 |
| Screenshot diffing | GPT-6 Luna | $10 | 6.9 | $6.24 | $0.69 | 1h 32m | 354 |
| Screenshot diffing | GPT-6 Luna | $50 | 4.6 | $13.96 | $0.85 | 1h 50m | 429 |
| Screenshot diffing | GPT-6 Luna | $200 | 2.9 | $25.47 | $1.21 | 3h 19m | 742 |
| Screenshot diffing | Gemini 3.8 Flash | $10 | 0.0 | $0.14 | $9.35 | 1h 7m | 684 |
| Screenshot diffing | Gemini 3.8 Flash | $50 | 0.0 | $0.01 | $7.11 | 1h 2m | 570 |
| Screenshot diffing | Gemini 3.8 Flash | $200 | 0.0 | $0.05 | $7.64 | 1h 17m | 495 |
| Screenshot diffing | Grok 4.7 | $10 | 0.0 | $6.92 | $29.04 | 2h 14m | 59 |
| Screenshot diffing | Grok 4.7 | $50 | 8.8 | $10.48 | $26.34 | 1h 55m | 155 |
| Screenshot diffing | Grok 4.7 | $200 | 8.0 | $13.97 | $34.78 | 2h 31m | 254 |
| Screenshot diffing | Muse Spark 1.3 | $10 | 7.9 | $3.60 | $8.17 | 4h 28m | 206 |
| Screenshot diffing | Muse Spark 1.3 | $50 | 3.4 | $9.00 | $10.79 | 5h 56m | 267 |
| Screenshot diffing | Muse Spark 1.3 | $200 | 12.7 | $45.99 | $13.21 | 4h 23m | 285 |
| Screenshot diffing | Muse Wonder | $10 | 13.0 | $4.06 | $18.12 | 4h 35m | 265 |
| Screenshot diffing | Muse Wonder | $50 | 1.6 | $5.41 | $16.91 | 8h 30m | 269 |
| Screenshot diffing | Muse Wonder | $200 | 6.5 | $42.06 | $25.84 | 8h 34m | 363 |
| Screenshot diffing | Kimi K3 | $10 | 17.1 | $9.62 | $16.36 | 5h 35m | 288 |
| Screenshot diffing | Kimi K3 | $50 | 19.6 | $44.10 | $24.14 | 9h 47m | 329 |
| Screenshot diffing | Kimi K3 | $200 | 0.0 | $82.30 | $62.72 | 10h 35m | 537 |
| Screenshot diffing | Qwen3.8 Max | $10 | 9.2 | $9.61 | $17.56 | 9h 31m | 418 |
| Screenshot diffing | Qwen3.8 Max | $50 | 2.9 | $28.33 | $14.64 | 6h 22m | 347 |
| Screenshot diffing | Qwen3.8 Max | $200 | 8.7 | $109.83 | $20.13 | 8h 55m | 438 |
| Screenshot diffing | GLM 5.3 | $10 | 11.6 | $9.47 | $18.79 | 8h 22m | 922 |
| Screenshot diffing | GLM 5.3 | $50 | 0.0 | $34.98 | $12.70 | 7h 55m | 572 |
| Screenshot diffing | GLM 5.3 | $200 | 0.0 | $144.70 | $10.71 | 7h 43m | 566 |
| Screenshot generation | Claude Opus 5.5 | $10 | 74.2 | $8.55 | $78.35 | 7h 3m | 273 |
| Screenshot generation | Claude Opus 5.5 | $50 | 63.0 | $19.37 | $40.51 | 4h 50m | 408 |
| Screenshot generation | Claude Opus 5.5 | $200 | 78.9 | $50.15 | $65.90 | 5h 51m | 352 |
| Screenshot generation | Claude Sonnet 5.5 | $10 | 57.8 | $7.80 | $36.38 | 7h 42m | 811 |
| Screenshot generation | Claude Sonnet 5.5 | $50 | 59.7 | $30.73 | $38.57 | 8h 53m | 518 |
| Screenshot generation | Claude Sonnet 5.5 | $200 | 64.6 | $58.14 | $44.26 | 10h 35m | 936 |
| Screenshot generation | Claude Fable 5.1 | $10 | 44.6 | $5.56 | $47.20 | 2h 46m | 226 |
| Screenshot generation | Claude Fable 5.1 | $50 | 0.0 | $19.00 | $45.56 | 2h 34m | 254 |
| Screenshot generation | Claude Fable 5.1 | $200 | 35.0 | $32.00 | $46.10 | 2h 55m | 221 |
| Screenshot generation | Claude Haiku 5.5 | $10 | 35.9 | $6.75 | $5.62 | 3h 31m | 286 |
| Screenshot generation | Claude Haiku 5.5 | $50 | 21.6 | $5.82 | $5.07 | 1h 53m | 244 |
| Screenshot generation | Claude Haiku 5.5 | $200 | 40.6 | $14.33 | $3.66 | 2h 50m | 202 |
| Screenshot generation | GPT-6 Astra | $10 | 47.3 | $6.11 | $43.74 | 1h 17m | 222 |
| Screenshot generation | GPT-6 Astra | $50 | 59.3 | $22.70 | $163.70 | 5h 47m | 785 |
| Screenshot generation | GPT-6 Astra | $200 | 50.6 | $22.32 | $36.47 | 1h 18m | 196 |
| Screenshot generation | GPT-6.1 Sol | $10 | 30.1 | $7.28 | $8.09 | 2h 17m | 282 |
| Screenshot generation | GPT-6.1 Sol | $50 | 40.5 | $13.98 | $5.16 | 1h 42m | 198 |
| Screenshot generation | GPT-6.1 Sol | $200 | 42.1 | $14.01 | $9.78 | 3h 8m | 347 |
| Screenshot generation | GPT-6 Luna | $10 | 14.1 | $0.00 | $0.06 | 10m | 43 |
| Screenshot generation | GPT-6 Luna | $50 | 16.5 | $0.00 | $0.14 | 20m | 87 |
| Screenshot generation | GPT-6 Luna | $200 | 21.0 | $0.00 | $0.10 | 23m | 84 |
| Screenshot generation | Gemini 3.8 Flash | $10 | 8.6 | $0.02 | $13.03 | 59m | 888 |
| Screenshot generation | Gemini 3.8 Flash | $50 | 16.2 | $0.32 | $6.52 | 40m | 494 |
| Screenshot generation | Gemini 3.8 Flash | $200 | 0.0 | $0.47 | $7.54 | 1h 5m | 547 |
| Screenshot generation | Grok 4.7 | $10 | 28.6 | $7.02 | $20.24 | 1h 14m | 96 |
| Screenshot generation | Grok 4.7 | $50 | 21.1 | $11.39 | $16.04 | 1h 4m | 74 |
| Screenshot generation | Grok 4.7 | $200 | 17.9 | $27.93 | $30.09 | 1h 56m | 158 |
| Screenshot generation | Muse Spark 1.3 | $10 | 16.8 | $3.62 | $22.39 | 4h 5m | 435 |
| Screenshot generation | Muse Spark 1.3 | $50 | 0.0 | $5.24 | $11.49 | 2h 44m | 270 |
| Screenshot generation | Muse Spark 1.3 | $200 | 19.8 | $7.53 | $28.81 | 5h 10m | 494 |
| Screenshot generation | Muse Wonder | $10 | 32.5 | $1.71 | $17.26 | 3h 17m | 232 |
| Screenshot generation | Muse Wonder | $50 | 22.4 | $14.03 | $13.50 | 2h 20m | 175 |
| Screenshot generation | Muse Wonder | $200 | 18.6 | $18.92 | $22.81 | 3h 34m | 273 |
| Screenshot generation | Kimi K3 | $10 | 32.7 | $4.86 | $97.78 | 7h 35m | 583 |
| Screenshot generation | Kimi K3 | $50 | 0.0 | $18.13 | $63.57 | 9h 2m | 480 |
| Screenshot generation | Kimi K3 | $200 | 35.2 | $7.69 | $74.52 | 6h 40m | 588 |
| Screenshot generation | Qwen3.8 Max | $10 | 24.9 | $0.90 | $7.96 | 2h 51m | 238 |
| Screenshot generation | Qwen3.8 Max | $50 | 29.6 | $1.01 | $7.76 | 3h 13m | 225 |
| Screenshot generation | Qwen3.8 Max | $200 | 32.1 | $4.83 | $12.44 | 3h 22m | 444 |
| Screenshot generation | GLM 5.3 | $10 | 0.0 | $1.86 | $24.93 | 7h 57m | 1183 |
| Screenshot generation | GLM 5.3 | $50 | 0.0 | $14.07 | $66.21 | 5h 28m | 958 |
| Screenshot generation | GLM 5.3 | $200 | 32.7 | $1.94 | $124.79 | 7h 52m | 1692 |
| Speculative drafting | Claude Opus 5.5 | $10 | 21.2 | $9.38 | $40.30 | 4h 38m | 204 |
| Speculative drafting | Claude Opus 5.5 | $50 | 37.0 | $39.11 | $64.38 | 7h 10m | 223 |
| Speculative drafting | Claude Opus 5.5 | $200 | 41.8 | $175.20 | $56.64 | 7h 10m | 228 |
| Speculative drafting | Claude Sonnet 5.5 | $10 | 20.8 | $9.90 | $22.77 | 6h 48m | 265 |
| Speculative drafting | Claude Sonnet 5.5 | $50 | 33.8 | $45.53 | $29.41 | 7h 49m | 389 |
| Speculative drafting | Claude Sonnet 5.5 | $200 | 48.4 | $176.61 | $41.93 | 9h 26m | 483 |
| Speculative drafting | Claude Fable 5.1 | $10 | 15.6 | $9.90 | $105.27 | 5h 59m | 108 |
| Speculative drafting | Claude Fable 5.1 | $50 | 26.8 | $31.41 | $84.54 | 5h 16m | 116 |
| Speculative drafting | Claude Fable 5.1 | $200 | 29.5 | $82.67 | $199.70 | 9h 7m | 200 |
| Speculative drafting | Claude Haiku 5.5 | $10 | 5.5 | $6.18 | $7.56 | 4h 56m | 344 |
| Speculative drafting | Claude Haiku 5.5 | $50 | 18.6 | $19.20 | $5.65 | 4h 22m | 282 |
| Speculative drafting | Claude Haiku 5.5 | $200 | 21.3 | $47.47 | $9.59 | 7h 3m | 435 |
| Speculative drafting | GPT-6 Astra | $10 | 16.8 | $9.89 | $64.98 | 2h 57m | 341 |
| Speculative drafting | GPT-6 Astra | $50 | 16.5 | $36.16 | $150.72 | 7h 55m | 752 |
| Speculative drafting | GPT-6 Astra | $200 | 25.8 | $191.58 | $184.70 | 9h 48m | 895 |
| Speculative drafting | GPT-6.1 Sol | $10 | 8.1 | $8.63 | $5.94 | 3h 2m | 190 |
| Speculative drafting | GPT-6.1 Sol | $50 | 0.0 | $45.37 | $9.40 | 7h 56m | 266 |
| Speculative drafting | GPT-6.1 Sol | $200 | 25.2 | $150.84 | $10.69 | 10h 8m | 283 |
| Speculative drafting | GPT-6 Luna | $10 | 0.0 | $3.49 | $0.14 | 22m | 94 |
| Speculative drafting | GPT-6 Luna | $50 | 0.0 | $1.49 | $0.21 | 34m | 145 |
| Speculative drafting | GPT-6 Luna | $200 | 0.0 | $2.88 | $0.26 | 43m | 167 |
| Speculative drafting | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $5.41 | 34m | 435 |
| Speculative drafting | Gemini 3.8 Flash | $50 | 0.0 | $2.98 | $6.12 | 1h 23m | 409 |
| Speculative drafting | Gemini 3.8 Flash | $200 | 0.0 | $8.73 | $17.98 | 2h 6m | 1290 |
| Speculative drafting | Grok 4.7 | $10 | 0.0 | $9.75 | $30.70 | 2h 52m | 76 |
| Speculative drafting | Grok 4.7 | $50 | 0.0 | $34.90 | $33.77 | 3h 2m | 123 |
| Speculative drafting | Grok 4.7 | $200 | 0.0 | $22.28 | $61.31 | 5h 51m | 62 |
| Speculative drafting | Muse Spark 1.3 | $10 | 0.0 | $4.50 | $11.61 | 9h 49m | 272 |
| Speculative drafting | Muse Spark 1.3 | $50 | 0.0 | $34.52 | $15.13 | 8h 50m | 307 |
| Speculative drafting | Muse Spark 1.3 | $200 | 8.7 | $34.52 | $13.15 | 6h 49m | 270 |
| Speculative drafting | Muse Wonder | $10 | 9.5 | $8.91 | $12.17 | 4h 34m | 189 |
| Speculative drafting | Muse Wonder | $50 | 13.1 | $38.39 | $15.86 | 5h 17m | 228 |
| Speculative drafting | Muse Wonder | $200 | 5.2 | $34.52 | $14.64 | 9h 11m | 230 |
| Speculative drafting | Kimi K3 | $10 | 0.0 | $9.99 | $11.04 | 4h 15m | 197 |
| Speculative drafting | Kimi K3 | $50 | 9.9 | $34.52 | $18.90 | 9h 11m | 302 |
| Speculative drafting | Kimi K3 | $200 | 0.0 | $38.39 | $27.92 | 10h 30m | 398 |
| Speculative drafting | Qwen3.8 Max | $10 | 0.0 | $9.06 | $13.29 | 7h 45m | 245 |
| Speculative drafting | Qwen3.8 Max | $50 | 0.0 | $22.20 | $18.61 | 10h 19m | 288 |
| Speculative drafting | Qwen3.8 Max | $200 | 7.0 | $143.21 | $13.88 | 5h 47m | 285 |
| Speculative drafting | GLM 5.3 | $10 | 0.0 | $9.99 | $12.18 | 4h 44m | 535 |
| Speculative drafting | GLM 5.3 | $50 | 12.0 | $32.32 | $10.27 | 8h 46m | 502 |
| Speculative drafting | GLM 5.3 | $200 | 13.7 | $31.65 | $26.78 | 8h 18m | 354 |
| SVG generation | Claude Opus 5.5 | $10 | 0.0 | $9.88 | $23.67 | 3h 28m | 178 |
| SVG generation | Claude Opus 5.5 | $50 | 5.2 | $45.64 | $23.12 | 5h 52m | 267 |
| SVG generation | Claude Opus 5.5 | $200 | 18.0 | $179.04 | $67.37 | 9h 32m | 312 |
| SVG generation | Claude Sonnet 5.5 | $10 | 0.0 | $9.79 | $21.79 | 4h 13m | 247 |
| SVG generation | Claude Sonnet 5.5 | $50 | 8.2 | $49.24 | $19.43 | 5h 25m | 450 |
| SVG generation | Claude Sonnet 5.5 | $200 | 13.9 | $183.81 | $29.58 | 8h 51m | 420 |
| SVG generation | Claude Fable 5.1 | $10 | 2.7 | $7.84 | $57.30 | 3h 17m | 153 |
| SVG generation | Claude Fable 5.1 | $50 | 2.0 | $49.82 | $191.52 | 9h 1m | 209 |
| SVG generation | Claude Fable 5.1 | $200 | 20.3 | $174.13 | $129.96 | 6h 42m | 153 |
| SVG generation | Claude Haiku 5.5 | $10 | 0.0 | $8.00 | $6.38 | 4h 1m | 213 |
| SVG generation | Claude Haiku 5.5 | $50 | 10.0 | $36.68 | $9.24 | 6h 10m | 288 |
| SVG generation | Claude Haiku 5.5 | $200 | 10.1 | $160.62 | $11.89 | 8h 1m | 514 |
| SVG generation | GPT-6 Astra | $10 | 0.0 | $10.00 | $66.29 | 2h 55m | 397 |
| SVG generation | GPT-6 Astra | $50 | 8.3 | $49.76 | $122.84 | 5h 44m | 660 |
| SVG generation | GPT-6 Astra | $200 | 14.0 | $199.93 | $188.71 | 8h 56m | 961 |
| SVG generation | GPT-6.1 Sol | $10 | 3.2 | $8.47 | $5.46 | 1h 50m | 189 |
| SVG generation | GPT-6.1 Sol | $50 | 7.2 | $49.95 | $13.56 | 5h 44m | 431 |
| SVG generation | GPT-6.1 Sol | $200 | 0.0 | $199.93 | $20.88 | 10h 1m | 664 |
| SVG generation | GPT-6 Luna | $10 | 0.0 | $6.69 | $0.86 | 1h 51m | 536 |
| SVG generation | GPT-6 Luna | $50 | 0.0 | $7.72 | $0.52 | 1h 8m | 344 |
| SVG generation | GPT-6 Luna | $200 | 0.0 | $35.20 | $2.07 | 5h 52m | 1281 |
| SVG generation | Gemini 3.8 Flash | $10 | 5.3 | $0.05 | $13.10 | 2h 18m | 946 |
| SVG generation | Gemini 3.8 Flash | $50 | 0.0 | $5.82 | $14.14 | 12h | 854 |
| SVG generation | Gemini 3.8 Flash | $200 | 0.0 | $8.73 | $6.69 | 1h 18m | 453 |
| SVG generation | Grok 4.7 | $10 | 0.0 | $9.41 | $19.17 | 2h 10m | 109 |
| SVG generation | Grok 4.7 | $50 | 0.0 | $15.68 | $46.79 | 9h 54m | 88 |
| SVG generation | Grok 4.7 | $200 | 0.0 | $57.53 | $28.39 | 4h 5m | 114 |
| SVG generation | Muse Spark 1.3 | $10 | 0.0 | $9.51 | $15.15 | 8h 1m | 295 |
| SVG generation | Muse Spark 1.3 | $50 | 0.0 | $23.27 | $6.26 | 4h 32m | 173 |
| SVG generation | Muse Spark 1.3 | $200 | 0.0 | $92.78 | $11.48 | 8h 28m | 254 |
| SVG generation | Muse Wonder | $10 | 0.0 | $8.91 | $18.62 | 6h 16m | 238 |
| SVG generation | Muse Wonder | $50 | 0.0 | $21.39 | $25.73 | 10h 1m | 319 |
| SVG generation | Muse Wonder | $200 | 0.0 | $35.87 | $7.70 | 4h 13m | 171 |
| SVG generation | Kimi K3 | $10 | 0.0 | $9.89 | $24.19 | 3h 10m | 195 |
| SVG generation | Kimi K3 | $50 | 0.0 | $48.45 | $40.22 | 8h 37m | 349 |
| SVG generation | Kimi K3 | $200 | 0.0 | $84.47 | $40.99 | 9h 59m | 373 |
| SVG generation | Qwen3.8 Max | $10 | 0.0 | $8.26 | $12.71 | 6h 37m | 266 |
| SVG generation | Qwen3.8 Max | $50 | 0.0 | $40.87 | $17.33 | 9h 41m | 330 |
| SVG generation | Qwen3.8 Max | $200 | 0.0 | $167.80 | $17.17 | 7h 10m | 360 |
| SVG generation | GLM 5.3 | $10 | 0.0 | $8.65 | $15.47 | 3h 18m | 243 |
| SVG generation | GLM 5.3 | $50 | 0.0 | $49.43 | $40.68 | 8h 43m | 543 |
| SVG generation | GLM 5.3 | $200 | 0.0 | $92.83 | $27.68 | 11h 40m | 441 |
| Ternary quantization | Claude Opus 5.5 | $10 | 22.7 | $8.49 | $36.53 | 4h 46m | 275 |
| Ternary quantization | Claude Opus 5.5 | $50 | 36.5 | $46.06 | $93.36 | 9h 57m | 284 |
| Ternary quantization | Claude Opus 5.5 | $200 | 49.1 | $195.44 | $26.90 | 7h 11m | 338 |
| Ternary quantization | Claude Sonnet 5.5 | $10 | 24.7 | $9.03 | $41.57 | 5h 43m | 807 |
| Ternary quantization | Claude Sonnet 5.5 | $50 | 18.9 | $35.89 | $42.04 | 10h 34m | 457 |
| Ternary quantization | Claude Sonnet 5.5 | $200 | 41.2 | $192.65 | $42.30 | 7h 46m | 790 |
| Ternary quantization | Claude Fable 5.1 | $10 | 7.7 | $9.83 | $46.91 | 3h 20m | 148 |
| Ternary quantization | Claude Fable 5.1 | $50 | 29.3 | $31.41 | $97.28 | 5h 46m | 200 |
| Ternary quantization | Claude Fable 5.1 | $200 | 28.9 | $73.29 | $263.42 | 9h 15m | 206 |
| Ternary quantization | Claude Haiku 5.5 | $10 | 8.7 | $8.73 | $4.21 | 2h 43m | 199 |
| Ternary quantization | Claude Haiku 5.5 | $50 | 1.2 | $34.32 | $14.53 | 7h 29m | 482 |
| Ternary quantization | Claude Haiku 5.5 | $200 | 7.9 | $106.34 | $22.37 | 9h 42m | 360 |
| Ternary quantization | GPT-6 Astra | $10 | 1.3 | $9.89 | $58.53 | 2h 30m | 322 |
| Ternary quantization | GPT-6 Astra | $50 | 0.2 | $49.98 | $187.07 | 7h 23m | 863 |
| Ternary quantization | GPT-6 Astra | $200 | 19.5 | $136.11 | $177.91 | 10h 6m | 848 |
| Ternary quantization | GPT-6.1 Sol | $10 | 1.5 | $4.69 | $4.96 | 1h 23m | 191 |
| Ternary quantization | GPT-6.1 Sol | $50 | 2.1 | $38.39 | $10.91 | 8h 48m | 391 |
| Ternary quantization | GPT-6.1 Sol | $200 | 4.6 | $166.71 | $11.52 | 10h 33m | 316 |
| Ternary quantization | GPT-6 Luna | $10 | 0.0 | $5.76 | $0.54 | 58m | 296 |
| Ternary quantization | GPT-6 Luna | $50 | 0.0 | $3.49 | $0.33 | 58m | 162 |
| Ternary quantization | GPT-6 Luna | $200 | 0.1 | $8.63 | $0.93 | 2h 6m | 515 |
| Ternary quantization | Gemini 3.8 Flash | $10 | 0.4 | $0.00 | $10.98 | 1h 11m | 818 |
| Ternary quantization | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $9.85 | 58m | 710 |
| Ternary quantization | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $12.86 | 1h 48m | 1031 |
| Ternary quantization | Grok 4.7 | $10 | 0.5 | $8.09 | $37.24 | 2h 41m | 102 |
| Ternary quantization | Grok 4.7 | $50 | 0.2 | $17.45 | $52.85 | 4h 35m | 179 |
| Ternary quantization | Grok 4.7 | $200 | 0.0 | $14.85 | $32.60 | 3h 25m | 197 |
| Ternary quantization | Muse Spark 1.3 | $10 | 0.0 | $0.00 | $15.06 | 7h 5m | 325 |
| Ternary quantization | Muse Spark 1.3 | $50 | 0.5 | $38.33 | $16.63 | 9h 39m | 326 |
| Ternary quantization | Muse Spark 1.3 | $200 | 9.3 | $70.84 | $20.63 | 11h 50m | 359 |
| Ternary quantization | Muse Wonder | $10 | 0.0 | $4.50 | $36.36 | 10h 16m | 474 |
| Ternary quantization | Muse Wonder | $50 | 0.0 | $17.82 | $24.12 | 10h 1m | 306 |
| Ternary quantization | Muse Wonder | $200 | 0.2 | $73.29 | $26.18 | 8h 5m | 287 |
| Ternary quantization | Kimi K3 | $10 | 5.5 | $9.99 | $31.57 | 8h 19m | 371 |
| Ternary quantization | Kimi K3 | $50 | 0.0 | $40.04 | $64.39 | 11h 25m | 634 |
| Ternary quantization | Kimi K3 | $200 | 1.4 | $60.41 | $29.46 | 7h 5m | 356 |
| Ternary quantization | Qwen3.8 Max | $10 | 2.5 | $8.91 | $36.84 | 6h | 375 |
| Ternary quantization | Qwen3.8 Max | $50 | 5.4 | $49.45 | $12.59 | 5h 53m | 333 |
| Ternary quantization | Qwen3.8 Max | $200 | 3.7 | $41.88 | $38.65 | 9h 3m | 470 |
| Ternary quantization | GLM 5.3 | $10 | 0.0 | $9.89 | $7.44 | 4h 31m | 274 |
| Ternary quantization | GLM 5.3 | $50 | 10.1 | $38.39 | $37.38 | 6h 31m | 520 |
| Ternary quantization | GLM 5.3 | $200 | 2.3 | $71.26 | $61.71 | 9h 20m | 831 |
| Video frame interpolation | Claude Opus 5.5 | $10 | 91.1 | $8.10 | $57.94 | 7h 58m | 294 |
| Video frame interpolation | Claude Opus 5.5 | $50 | 86.7 | $41.88 | $23.07 | 5h 15m | 289 |
| Video frame interpolation | Claude Opus 5.5 | $200 | 87.8 | $101.21 | $82.47 | 9h 50m | 259 |
| Video frame interpolation | Claude Sonnet 5.5 | $10 | 86.0 | $8.78 | $28.75 | 10h 3m | 318 |
| Video frame interpolation | Claude Sonnet 5.5 | $50 | 86.3 | $42.90 | $33.76 | 7h 54m | 378 |
| Video frame interpolation | Claude Sonnet 5.5 | $200 | 93.2 | $130.68 | $34.51 | 9h 48m | 300 |
| Video frame interpolation | Claude Fable 5.1 | $10 | 78.5 | $4.50 | $99.40 | 5h 53m | 107 |
| Video frame interpolation | Claude Fable 5.1 | $50 | 81.8 | $47.70 | $119.28 | 6h 40m | 174 |
| Video frame interpolation | Claude Fable 5.1 | $200 | 58.2 | $59.33 | $130.88 | 6h 43m | 153 |
| Video frame interpolation | Claude Haiku 5.5 | $10 | 63.7 | $7.43 | $6.51 | 5h 11m | 332 |
| Video frame interpolation | Claude Haiku 5.5 | $50 | 64.3 | $24.43 | $12.02 | 6h 17m | 268 |
| Video frame interpolation | Claude Haiku 5.5 | $200 | 65.4 | $28.77 | $14.78 | 7h 52m | 422 |
| Video frame interpolation | GPT-6 Astra | $10 | 85.3 | $10.00 | $268.79 | 10h 14m | 1167 |
| Video frame interpolation | GPT-6 Astra | $50 | 91.2 | $49.96 | $246.86 | 8h 45m | 1068 |
| Video frame interpolation | GPT-6 Astra | $200 | 84.6 | $143.31 | $276.33 | 11h 27m | 1157 |
| Video frame interpolation | GPT-6.1 Sol | $10 | 87.8 | $9.75 | $17.97 | 10h 1m | 454 |
| Video frame interpolation | GPT-6.1 Sol | $50 | 72.0 | $31.65 | $16.17 | 10h 26m | 478 |
| Video frame interpolation | GPT-6.1 Sol | $200 | 86.5 | $149.67 | $21.19 | 9h 56m | 600 |
| Video frame interpolation | GPT-6 Luna | $10 | 0.0 | $6.99 | $0.90 | 1h 19m | 469 |
| Video frame interpolation | GPT-6 Luna | $50 | 0.0 | $5.40 | $1.26 | 2h 54m | 663 |
| Video frame interpolation | GPT-6 Luna | $200 | 25.7 | $83.76 | $1.89 | 4h 49m | 1044 |
| Video frame interpolation | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $5.49 | 30m | 397 |
| Video frame interpolation | Gemini 3.8 Flash | $50 | 7.1 | $0.00 | $2.25 | 20m | 211 |
| Video frame interpolation | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $3.43 | 33m | 263 |
| Video frame interpolation | Grok 4.7 | $10 | 0.0 | $1.35 | $13.29 | 1h 12m | 171 |
| Video frame interpolation | Grok 4.7 | $50 | 0.0 | $17.45 | $21.17 | 1h 43m | 181 |
| Video frame interpolation | Grok 4.7 | $200 | 0.0 | $2.70 | $13.69 | 52m | 188 |
| Video frame interpolation | Muse Spark 1.3 | $10 | 23.7 | $5.40 | $9.61 | 3h 21m | 242 |
| Video frame interpolation | Muse Spark 1.3 | $50 | 49.8 | $4.50 | $14.20 | 6h 8m | 320 |
| Video frame interpolation | Muse Spark 1.3 | $200 | 31.2 | $66.11 | $11.28 | 5h 59m | 227 |
| Video frame interpolation | Muse Wonder | $10 | 68.8 | $5.85 | $19.94 | 5h 9m | 286 |
| Video frame interpolation | Muse Wonder | $50 | 49.8 | $5.40 | $12.73 | 5h 26m | 185 |
| Video frame interpolation | Muse Wonder | $200 | 55.9 | $31.21 | $10.82 | 3h 41m | 206 |
| Video frame interpolation | Kimi K3 | $10 | 0.0 | $6.30 | $36.78 | 8h 51m | 477 |
| Video frame interpolation | Kimi K3 | $50 | 32.5 | $40.14 | $35.64 | 6h 14m | 282 |
| Video frame interpolation | Kimi K3 | $200 | 79.5 | $69.80 | $23.91 | 5h 30m | 330 |
| Video frame interpolation | Qwen3.8 Max | $10 | 31.2 | $4.50 | $18.03 | 8h 52m | 286 |
| Video frame interpolation | Qwen3.8 Max | $50 | 0.0 | $13.23 | $13.86 | 6h 23m | 275 |
| Video frame interpolation | Qwen3.8 Max | $200 | 0.0 | $64.80 | $17.37 | 9h 17m | 274 |
| Video frame interpolation | GLM 5.3 | $10 | 26.9 | $4.05 | $45.90 | 6h 59m | 672 |
| Video frame interpolation | GLM 5.3 | $50 | 45.7 | $7.81 | $35.58 | 8h 41m | 492 |
| Video frame interpolation | GLM 5.3 | $200 | 52.3 | $33.57 | $9.74 | 9h 24m | 488 |
| Vision for a ternary LM | Claude Opus 5.5 | $10 | 0.4 | $4.08 | $25.76 | 2h 33m | 254 |
| Vision for a ternary LM | Claude Opus 5.5 | $50 | 6.5 | $26.39 | $54.88 | 5h 17m | 202 |
| Vision for a ternary LM | Claude Opus 5.5 | $200 | 2.4 | $111.68 | $43.58 | 6h 58m | 414 |
| Vision for a ternary LM | Claude Sonnet 5.5 | $10 | 0.0 | $5.41 | $17.28 | 2h 37m | 312 |
| Vision for a ternary LM | Claude Sonnet 5.5 | $50 | 0.0 | $18.43 | $21.56 | 3h 35m | 342 |
| Vision for a ternary LM | Claude Sonnet 5.5 | $200 | 0.0 | $174.17 | $46.67 | 7h 7m | 667 |
| Vision for a ternary LM | Claude Fable 5.1 | $10 | 0.0 | $7.20 | $62.00 | 2h 51m | 74 |
| Vision for a ternary LM | Claude Fable 5.1 | $50 | 0.0 | $14.42 | $48.86 | 2h 55m | 135 |
| Vision for a ternary LM | Claude Fable 5.1 | $200 | 0.0 | $52.35 | $68.56 | 4h 8m | 161 |
| Vision for a ternary LM | Claude Haiku 5.5 | $10 | 0.0 | $7.20 | $5.14 | 2h 24m | 230 |
| Vision for a ternary LM | Claude Haiku 5.5 | $50 | 0.0 | $14.39 | $7.15 | 3h 13m | 194 |
| Vision for a ternary LM | Claude Haiku 5.5 | $200 | 0.0 | $48.86 | $8.11 | 3h 53m | 310 |
| Vision for a ternary LM | GPT-6 Astra | $10 | 0.0 | $8.91 | $122.17 | 5h 17m | 548 |
| Vision for a ternary LM | GPT-6 Astra | $50 | 0.0 | $49.24 | $209.87 | 8h 7m | 999 |
| Vision for a ternary LM | GPT-6 Astra | $200 | 0.0 | $101.58 | $131.38 | 5h 13m | 592 |
| Vision for a ternary LM | GPT-6.1 Sol | $10 | 0.0 | $9.96 | $7.78 | 3h 6m | 238 |
| Vision for a ternary LM | GPT-6.1 Sol | $50 | 0.0 | $43.47 | $16.93 | 7h 41m | 501 |
| Vision for a ternary LM | GPT-6.1 Sol | $200 | 0.0 | $112.83 | $13.77 | 10h 19m | 400 |
| Vision for a ternary LM | GPT-6 Luna | $10 | 0.0 | $8.15 | $0.67 | 1h 46m | 369 |
| Vision for a ternary LM | GPT-6 Luna | $50 | 0.0 | $37.81 | $1.99 | 9h | 1178 |
| Vision for a ternary LM | GPT-6 Luna | $200 | 0.0 | $34.90 | $0.58 | 1h 11m | 307 |
| Vision for a ternary LM | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $18.82 | 2h 33m | 1183 |
| Vision for a ternary LM | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $13.45 | 1h 43m | 933 |
| Vision for a ternary LM | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $10.43 | 2h 12m | 598 |
| Vision for a ternary LM | Grok 4.7 | $10 | 0.0 | $8.78 | $30.41 | 2h 22m | 168 |
| Vision for a ternary LM | Grok 4.7 | $50 | 0.0 | $8.91 | $38.27 | 3h 40m | 61 |
| Vision for a ternary LM | Grok 4.7 | $200 | 0.0 | $27.92 | $33.57 | 2h 37m | 200 |
| Vision for a ternary LM | Muse Spark 1.3 | $10 | 0.0 | $9.99 | $13.72 | 6h 39m | 276 |
| Vision for a ternary LM | Muse Spark 1.3 | $50 | 0.0 | $34.52 | $19.60 | 11h 35m | 393 |
| Vision for a ternary LM | Muse Spark 1.3 | $200 | 0.0 | $114.60 | $29.09 | 11h 49m | 558 |
| Vision for a ternary LM | Muse Wonder | $10 | 0.0 | $9.66 | $19.82 | 5h 29m | 245 |
| Vision for a ternary LM | Muse Wonder | $50 | 0.0 | $31.41 | $18.17 | 6h 5m | 216 |
| Vision for a ternary LM | Muse Wonder | $200 | 0.0 | $31.65 | $15.25 | 9h 55m | 234 |
| Vision for a ternary LM | Kimi K3 | $10 | 0.0 | $8.63 | $13.68 | 2h 52m | 176 |
| Vision for a ternary LM | Kimi K3 | $50 | 0.0 | $27.99 | $23.94 | 7h 1m | 333 |
| Vision for a ternary LM | Kimi K3 | $200 | 0.0 | $41.93 | $79.95 | 8h 7m | 637 |
| Vision for a ternary LM | Qwen3.8 Max | $10 | 0.0 | $9.83 | $14.56 | 6h 26m | 296 |
| Vision for a ternary LM | Qwen3.8 Max | $50 | 0.0 | $27.92 | $9.24 | 4h 22m | 230 |
| Vision for a ternary LM | Qwen3.8 Max | $200 | 0.0 | $31.66 | $21.65 | 9h 55m | 464 |
| Vision for a ternary LM | GLM 5.3 | $10 | 0.0 | $8.63 | $27.17 | 3h 6m | 233 |
| Vision for a ternary LM | GLM 5.3 | $50 | 0.0 | $18.83 | $49.67 | 8h 16m | 632 |
| Vision for a ternary LM | GLM 5.3 | $200 | 0.0 | $31.65 | $13.26 | 10h 42m | 601 |
| Voice endpointing | Claude Opus 5.5 | $10 | 74.7 | $1.80 | $27.01 | 3h 14m | 210 |
| Voice endpointing | Claude Opus 5.5 | $50 | 68.8 | $6.73 | $52.64 | 4h 59m | 250 |
| Voice endpointing | Claude Opus 5.5 | $200 | 55.8 | $34.90 | $28.77 | 5h 3m | 336 |
| Voice endpointing | Claude Sonnet 5.5 | $10 | 49.4 | $0.98 | $12.40 | 1h 41m | 241 |
| Voice endpointing | Claude Sonnet 5.5 | $50 | 46.1 | $5.40 | $21.54 | 4h 41m | 439 |
| Voice endpointing | Claude Sonnet 5.5 | $200 | 57.4 | $5.94 | $16.52 | 3h 36m | 319 |
| Voice endpointing | Claude Fable 5.1 | $10 | 65.0 | $3.90 | $83.07 | 4h 40m | 137 |
| Voice endpointing | Claude Fable 5.1 | $50 | 0.0 | $2.70 | $45.10 | 3h 1m | 152 |
| Voice endpointing | Claude Fable 5.1 | $200 | 50.7 | $27.92 | $24.56 | 1h 47m | 107 |
| Voice endpointing | Claude Haiku 5.5 | $10 | 33.6 | $0.00 | $7.75 | 4h 11m | 274 |
| Voice endpointing | Claude Haiku 5.5 | $50 | 3.7 | $20.94 | $7.31 | 3h 49m | 312 |
| Voice endpointing | Claude Haiku 5.5 | $200 | 19.9 | $20.94 | $13.59 | 6h 19m | 249 |
| Voice endpointing | GPT-6 Astra | $10 | 50.5 | $1.80 | $27.60 | 56m | 171 |
| Voice endpointing | GPT-6 Astra | $50 | 62.9 | $13.96 | $20.62 | 37m | 108 |
| Voice endpointing | GPT-6 Astra | $200 | 60.4 | $10.47 | $24.68 | 50m | 119 |
| Voice endpointing | GPT-6.1 Sol | $10 | 45.8 | $1.80 | $3.82 | 1h 18m | 161 |
| Voice endpointing | GPT-6.1 Sol | $50 | 0.0 | $13.96 | $3.35 | 1h 5m | 120 |
| Voice endpointing | GPT-6.1 Sol | $200 | 42.3 | $37.99 | $7.49 | 2h 26m | 241 |
| Voice endpointing | GPT-6 Luna | $10 | 29.0 | $0.00 | $0.17 | 16m | 119 |
| Voice endpointing | GPT-6 Luna | $50 | 32.4 | $0.00 | $0.12 | 18m | 71 |
| Voice endpointing | GPT-6 Luna | $200 | 43.7 | $0.00 | $0.47 | 1h | 251 |
| Voice endpointing | Gemini 3.8 Flash | $10 | 40.0 | $0.00 | $8.17 | 24m | 510 |
| Voice endpointing | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $4.96 | 41m | 364 |
| Voice endpointing | Gemini 3.8 Flash | $200 | 0.0 | $0.00 | $4.95 | 1h 21m | 290 |
| Voice endpointing | Grok 4.7 | $10 | 43.3 | $2.97 | $10.96 | 50m | 147 |
| Voice endpointing | Grok 4.7 | $50 | 0.0 | $24.43 | $21.39 | 1h 40m | 133 |
| Voice endpointing | Grok 4.7 | $200 | 0.0 | $1.80 | $15.46 | 1h 17m | 54 |
| Voice endpointing | Muse Spark 1.3 | $10 | 34.1 | $5.40 | $17.33 | 4h 32m | 309 |
| Voice endpointing | Muse Spark 1.3 | $50 | 0.0 | $3.60 | $10.19 | 2h 59m | 224 |
| Voice endpointing | Muse Spark 1.3 | $200 | 0.0 | $6.75 | $17.35 | 7h 49m | 339 |
| Voice endpointing | Muse Wonder | $10 | 0.0 | $1.79 | $16.09 | 2h 57m | 189 |
| Voice endpointing | Muse Wonder | $50 | 5.2 | $4.05 | $19.98 | 3h 41m | 250 |
| Voice endpointing | Muse Wonder | $200 | 0.0 | $4.05 | $19.33 | 5h 56m | 243 |
| Voice endpointing | Kimi K3 | $10 | 0.0 | $4.95 | $33.48 | 8h 23m | 365 |
| Voice endpointing | Kimi K3 | $50 | 0.0 | $18.73 | $20.80 | 4h 55m | 273 |
| Voice endpointing | Kimi K3 | $200 | 0.0 | $47.61 | $50.87 | 9h 31m | 529 |
| Voice endpointing | Qwen3.8 Max | $10 | 50.5 | $3.60 | $11.14 | 5h 37m | 308 |
| Voice endpointing | Qwen3.8 Max | $50 | 33.2 | $15.60 | $13.77 | 6h 26m | 263 |
| Voice endpointing | Qwen3.8 Max | $200 | 0.0 | $11.88 | $18.47 | 5h 27m | 319 |
| Voice endpointing | GLM 5.3 | $10 | 40.8 | $4.13 | $10.67 | 7h 9m | 502 |
| Voice endpointing | GLM 5.3 | $50 | 0.0 | $1.03 | $12.22 | 7h 37m | 606 |
| Voice endpointing | GLM 5.3 | $200 | 0.0 | $33.08 | $21.58 | 8h 3m | 615 |
| Zero-shot classification | Claude Opus 5.5 | $10 | 59.3 | $8.95 | $16.20 | 3h 13m | 191 |
| Zero-shot classification | Claude Opus 5.5 | $50 | 57.7 | $14.11 | $33.36 | 3h 32m | 208 |
| Zero-shot classification | Claude Opus 5.5 | $200 | 59.0 | $47.38 | $56.42 | 5h 42m | 369 |
| Zero-shot classification | Claude Sonnet 5.5 | $10 | 62.0 | $6.99 | $24.63 | 3h 22m | 337 |
| Zero-shot classification | Claude Sonnet 5.5 | $50 | 63.6 | $34.48 | $20.30 | 5h 55m | 430 |
| Zero-shot classification | Claude Sonnet 5.5 | $200 | 0.0 | $87.99 | $50.93 | 7h 20m | 649 |
| Zero-shot classification | Claude Fable 5.1 | $10 | 59.6 | $9.60 | $91.76 | 5h 2m | 115 |
| Zero-shot classification | Claude Fable 5.1 | $50 | 0.0 | $44.98 | $162.48 | 8h 2m | 140 |
| Zero-shot classification | Claude Fable 5.1 | $200 | 62.1 | $47.64 | $76.74 | 5h | 182 |
| Zero-shot classification | Claude Haiku 5.5 | $10 | 52.3 | $5.88 | $6.81 | 3h 9m | 272 |
| Zero-shot classification | Claude Haiku 5.5 | $50 | 57.4 | $23.01 | $8.76 | 4h 57m | 368 |
| Zero-shot classification | Claude Haiku 5.5 | $200 | 48.1 | $30.10 | $7.44 | 5h 20m | 330 |
| Zero-shot classification | GPT-6 Astra | $10 | 52.5 | $5.24 | $20.45 | 39m | 110 |
| Zero-shot classification | GPT-6 Astra | $50 | 47.8 | $11.51 | $33.99 | 1h 17m | 210 |
| Zero-shot classification | GPT-6 Astra | $200 | 46.7 | $13.96 | $18.69 | 46m | 98 |
| Zero-shot classification | GPT-6.1 Sol | $10 | 38.7 | $5.29 | $4.92 | 1h 52m | 184 |
| Zero-shot classification | GPT-6.1 Sol | $50 | 50.2 | $18.05 | $7.98 | 3h 22m | 176 |
| Zero-shot classification | GPT-6.1 Sol | $200 | 56.5 | $30.50 | $5.07 | 2h 46m | 129 |
| Zero-shot classification | GPT-6 Luna | $10 | 36.9 | $3.09 | $0.57 | 1h 52m | 387 |
| Zero-shot classification | GPT-6 Luna | $50 | 29.4 | $13.96 | $0.41 | 1h 23m | 208 |
| Zero-shot classification | GPT-6 Luna | $200 | 24.8 | $6.98 | $0.36 | 27m | 181 |
| Zero-shot classification | Gemini 3.8 Flash | $10 | 0.0 | $0.00 | $3.85 | 56m | 263 |
| Zero-shot classification | Gemini 3.8 Flash | $50 | 0.0 | $0.00 | $31.56 | 1h 46m | 2088 |
| Zero-shot classification | Gemini 3.8 Flash | $200 | 29.6 | $0.00 | $5.67 | 52m | 386 |
| Zero-shot classification | Grok 4.7 | $10 | 0.0 | $6.98 | $15.40 | 1h 19m | 207 |
| Zero-shot classification | Grok 4.7 | $50 | 31.0 | $5.94 | $15.20 | 1h 3m | 181 |
| Zero-shot classification | Grok 4.7 | $200 | 17.4 | $14.19 | $8.30 | 43m | 138 |
| Zero-shot classification | Muse Spark 1.3 | $10 | 26.7 | $7.27 | $13.32 | 4h 32m | 300 |
| Zero-shot classification | Muse Spark 1.3 | $50 | 0.0 | $24.70 | $8.25 | 5h 1m | 234 |
| Zero-shot classification | Muse Spark 1.3 | $200 | 43.2 | $9.82 | $6.95 | 1h 58m | 166 |
| Zero-shot classification | Muse Wonder | $10 | 48.6 | $2.25 | $11.28 | 2h 48m | 164 |
| Zero-shot classification | Muse Wonder | $50 | 16.0 | $17.45 | $5.24 | 2h | 129 |
| Zero-shot classification | Muse Wonder | $200 | 37.1 | $14.85 | $12.51 | 5h 56m | 180 |
| Zero-shot classification | Kimi K3 | $10 | 41.2 | $6.38 | $17.02 | 3h 17m | 245 |
| Zero-shot classification | Kimi K3 | $50 | 40.2 | $28.31 | $17.62 | 6h 15m | 267 |
| Zero-shot classification | Kimi K3 | $200 | 0.0 | $13.96 | $13.33 | 3h 42m | 222 |
| Zero-shot classification | Qwen3.8 Max | $10 | 0.0 | $9.90 | $13.81 | 7h 41m | 253 |
| Zero-shot classification | Qwen3.8 Max | $50 | 43.2 | $19.43 | $8.88 | 4h 56m | 272 |
| Zero-shot classification | Qwen3.8 Max | $200 | 40.4 | $19.16 | $14.18 | 8h 28m | 293 |
| Zero-shot classification | GLM 5.3 | $10 | 13.8 | $4.12 | $14.75 | 3h 10m | 204 |
| Zero-shot classification | GLM 5.3 | $50 | 41.1 | $28.48 | $29.18 | 5h 16m | 429 |
| Zero-shot classification | GLM 5.3 | $200 | 0.0 | $62.37 | $30.52 | 6h 17m | 466 |