Modality: llm_chat · full deep dive — every ranked model, test result, and artifact.
Ranked by confidence-adjusted score (single/zero-sample, non-curated scores floored; curated empirical scores trusted as-is).
| # | Model | Provider | Adj. Score | Raw | Evidence |
|---|---|---|---|---|---|
| 1 | claude-sonnet-4-6 | claude_desktop | 0.773 | 0.850 | n=50 |
| 2 | deepseek-chat | deepseek | 0.642 | 0.998 | n=9 |
| 3 | claude-sonnet-4-6::think=fast | claude_code | 0.545 | 1.000 | n=6 |
| 4 | claude-sonnet-4-6::think=deep | claude_code | 0.545 | 1.000 | n=6 |
| 5 | gpt-5.5::think=fast | codex | 0.525 | 0.963 | n=6 |
| 6 | gpt-5.5::think=deep | codex | 0.525 | 0.963 | n=6 |
| 7 | claude-opus-4-7::think=fast | claude_code | 0.477 | 0.875 | n=6 |
| 8 | claude-opus-4-7::think=deep | claude_code | 0.477 | 0.875 | n=6 |
| 9 | gpt-4o | openai | 0.456 | 0.456 | curated |
| 10 | claude-opus-4-8::think=fast | claude_code | 0.430 | 0.788 | n=6 |
| 11 | claude-opus-4-8::think=deep | claude_code | 0.382 | 0.700 | n=6 |
| 12 | gpt-5.3-codex-spark::think=deep | codex | 0.382 | 0.700 | n=6 |
| 13 | gemini-2.5-flash | google_gemini | 0.347 | 0.407 | n=29 |
| 14 | gpt-5.4::think=fast | codex | 0.334 | 0.612 | n=6 |
| 15 | gpt-5.4::think=deep | codex | 0.334 | 0.612 | n=6 |
| 16 | claude-sonnet-4-5-20250929 | anthropic | 0.332 | 0.332 | n=11550 |
| 17 | claude-sonnet-4-6 | anthropic | 0.292 | 0.292 | curated |
| 18 | gpt-5.3-codex-spark::think=fast | codex | 0.286 | 0.525 | n=6 |
| 19 | claude-haiku-4-5-20251001 | anthropic | 0.277 | 0.315 | n=37 |
| 20 | claude-opus-4-6 | anthropic | 0.239 | 0.239 | curated |
| 21 | gemini-3.1-pro | gemini | 0.120 | 0.800 | provisional (n≤1) |
| 22 | antigravity:gemini-2.5-pro | antigravity | 0.117 | 0.780 | provisional (n≤1) |
| 23 | antigravity:gemini-2.5-flash | antigravity | 0.105 | 0.700 | provisional (n≤1) |
| 24 | antigravity:gemini-2.0-flash-thinking | antigravity | 0.105 | 0.700 | provisional (n≤1) |
| 25 | gemini-2.5-pro | google_gemini | 0.017 | 0.017 | n=8006 |
| 26 | o3 | openai | 0.000 | 0.000 | provisional (n≤1) |
| 27 | o3-mini | openai | 0.000 | 0.000 | provisional (n≤1) |
| 28 | claude-sonnet-4-6 | claude_code | 0.000 | 0.000 | n=50 |
| 29 | claude-opus-4-7 | claude_code | 0.000 | 0.000 | n=50 |
| 30 | claude-opus-4-8 | claude_code | 0.000 | 0.000 | n=50 |
| 31 | claude-opus-4-6 | claude_code | 0.000 | 0.000 | n=50 |
| 32 | claude-opus-4-5-20251101 | claude_code | 0.000 | 0.000 | n=50 |
| 33 | claude-sonnet-4-5-20250929 | claude_code | 0.000 | 0.000 | n=50 |
| 34 | claude-opus-4-1-20250805 | claude_code | 0.000 | 0.000 | n=50 |
| 35 | claude-opus-4-20250514 | claude_code | 0.000 | 0.000 | n=50 |
| 36 | claude-sonnet-4-20250514 | claude_code | 0.000 | 0.000 | n=50 |
| 37 | gpt-5.4 | codex | 0.000 | 0.000 | n=50 |
| 38 | gpt-5.5 | codex | 0.000 | 0.000 | n=50 |
| 39 | gpt-5.3-codex-spark | codex | 0.000 | 0.000 | n=50 |
| 40 | copilot:gpt-4o | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 41 | copilot:gpt-4o-mini | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 42 | copilot:claude-3.5-sonnet | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 43 | copilot:claude-3.7-sonnet | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 44 | copilot:gemini-1.5-pro | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 45 | copilot:gemini-2.0-flash | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 46 | copilot:o1 | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
| 47 | copilot:o3-mini | github_copilot | 0.000 | 0.000 | provisional (n≤1) |
No benchmark outputs recorded for this niche yet.