Agentic coding
Resolving real repository and terminal tasks through a tool-using loop. Every score here measures a model together with a scaffold, so entries carry the scaffold as a variant and two rows for one model are expected rather than a duplicate.
Ordered by how many of these benchmarks put a model in their top 10, then by its best placing. Each column keeps its own ranking; nothing is averaged.
| Model | In top 10 | Artificial Analysis LLM Leaderboard Terminal-Bench 2.1 | Artificial Analysis Coding Agents Artificial Analysis Coding Agent Index | Artificial Analysis Coding Agents Terminal-Bench v4 | HarnessTax — SWE-bench Lite SWE-bench Lite | HarnessTax — Terminal-Bench 2.0 Terminal-Bench 2.0 |
|---|---|---|---|---|---|---|
 GPT 5.6 Sol openai | 5/5 | #3 89.5% | #10 0.546 | #10 37.4% | #3 77.8% | #1 83.3% |
 Claude Fable 5.1 anthropic | 3/5 | #1 91.4% | #5 0.622 | #3 57.6% | — | — |
| GPT 6 Astra openai | 3/5 | #2 89.9% | #6 0.616 | #5 55.6% | — | — |
| Claude Opus 5 anthropic | 3/5 | #4 89.1% | #7 0.597 | #7 54.5% | — | — |
| Claude Fable 5 anthropic | 2/5 | #13 84.6% | — | — | #1 97.8% | #3 75.6% |
| Claude Sonnet 5.5 anthropic | 2/5 | — | #1 0.684 | #1 66.2% | — | — |
| Claude Opus 4.8 anthropic | 2/5 | #14 84.6% | — | — | #2 88.9% | #5 72.2% |
| GPT 5.6 Luna openai | 2/5 | #20 80.9% | #16 0.432 | #17 14.6% | #7 55.6% | #2 76.7% |
| Claude Opus 5.5 anthropic | 2/5 | — | #2 0.660 | #2 63.1% | — | — |
 Gemini 4 Argon google | 2/5 | — | #3 0.638 | #4 56.1% | — | — |
 Kimi K3 kimi | 2/5 | #12 85.0% | #13 0.519 | #13 21.2% | #4 76.7% | #4 73.3% |
| GPT 6.1 Sol openai | 2/5 | — | #4 0.629 | #6 54.5% | — | — |
| Claude Sonnet 4.6 anthropic | 2/5 | #36 71.2% | — | — | #5 68.9% | #6 65.6% |
| Claude Haiku 4.5 anthropic | 2/5 | #77 44.2% | — | — | #6 60.0% | #7 47.8% |
| GPT 6 Sol openai | 2/5 | — | #8 0.567 | #8 43.4% | — | — |
 Qwen 3.8 Max alibaba | 1/5 | #5 88.8% | #15 0.433 | #15 16.7% | — | — |
 Grok 4.6 xai | 1/5 | #6 88.4% | #14 0.470 | #14 17.7% | — | — |
| GPT 5.6 Terra openai | 1/5 | #7 88.0% | — | — | — | — |
| Gemini 3.8 Flash google | 1/5 | #8 87.6% | #18 0.419 | #18 14.6% | — | — |
| Qwen 3.8 Flash Next alibaba | 1/5 | #9 86.1% | — | — | — | — |
 GLM 5.3 zai | 1/5 | #17 83.9% | #12 0.536 | #9 39.9% | — | — |
| Grok 4.7 xai | 1/5 | — | #9 0.563 | #11 33.3% | — | — |
| Gemini 3.7 Flash google | 1/5 | #10 85.8% | — | — | — | — |
A dash means the board does not list that model — not a score of zero. Entries not yet matched to a model are left out here; they still appear on their own board.