Overall (20-task cohort)
overall
13 runs · 13 models · evaluated by BixBench3 Score.
| # | Model ↕ | Variant ↕ | Mode ↕ | Score ↓ | Avg. dur ↕ | Tokens ↕ | Date ↕ |
|---|---|---|---|---|---|---|---|
| 1 | gpt-5.6-sol | — | max | 0.480 | 431.1m | 924.1M | 2026-08-26 |
| 2 | kimi-k3 | — | max | 0.470 | 405.0m | 560.7M | 2026-08-26 |
| 3 | glm-5.2 | — | max | 0.463 | 479.4m | 2736.1M | 2026-08-26 |
| 4 | claude-opus-4.8 | — | max | 0.460 | 431.5m | 1450.3M | 2026-08-26 |
| 5 | gpt-5.5 | — | max | 0.431 | 343.6m | 650.7M | 2026-08-26 |
| 6 | gemini-3.5-flash | — | max | 0.421 | 382.1m | 3588.4M | 2026-08-26 |
| 7 | claude-opus-5 | — | max | 0.406 | 332.3m | 1011.2M | 2026-08-26 |
| 8 | gpt-5.4-mini | — | max | 0.364 | 401.3m | 1511.6M | 2026-08-26 |
| 9 | gpt-5.4-nano | — | max | 0.275 | 322.9m | 2366.3M | 2026-08-26 |
| 10 | gemini-3.1-pro-preview | — | max | 0.248 | 556.0m | 5477.6M | 2026-08-26 |
| 11 | claude-sonnet-4.6 | — | max | 0.153 | 1087.6m | 5698.5M | 2026-08-26 |
| 12 | claude-haiku-4.5 | — | max | 0.015 | 51.5m | 333.4M | 2026-08-26 |
| 13 | gemini-3.1-flash-lite | — | max | 0.000 | 44.4m | 164.2M | 2026-08-26 |
Click column headers to sort. Click mode chips to filter.