RankingGPQA Diamond · source release snapshot; version not specified
GPQA Diamond · source release snapshot; version not specified
- Bucket
- Hard reasoning
- Unit
- percent
- Direction
- Higher is better
- Version
- source release snapshot; version not specified
- Display harness
- Qwen/Qwen3.8-2.4T-A95B
- Board
- https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B
Compare published benchmark results with category weights →
Models
26–29 of 29 entries
| Model | Score | Harness | Evidence | Source-recorded date |
|---|---|---|---|---|
| GPT-5.6 SolOpenAI | 94.1%Reported settings & sourceTable headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs. First-party reported result. moonshotai/Kimi-K3 · Performance table: GPQA Diamond · reviewed 2026-09-24 | moonshotai/Kimi-K3 | lab self-report | 2026-09-24 |
| Kimi K3Moonshot | 93.5%Reported settings & sourceTable headers: Kimi K3, GPT-5.6 Sol, Opus 4.8 and GLM-5.2 max; GPT-5.5 xhigh; Fable 5 max with fallbacks. Benchmark-specific footnotes override header settings; cited results are not new Moonshot evaluations. Benchmark-specific tools, harness and budget documented under Evaluation Details. Kimi temp1; single-step top_p.95,agentic top_p1; vision tools=Python,3runs except ZeroBench5runs. First-party reported result. moonshotai/Kimi-K3 · Performance table: GPQA Diamond · reviewed 2026-09-24 | moonshotai/Kimi-K3 | lab self-report | 2026-09-24 |
| Kimi-K2.6Moonshot | 90.5%Reported settings & sourceMAI avg 4 runs, temperature 1, top-p .97; simple ReAct bash/string-replace; Terminal-Bench ignores predefined timeouts. Comparator configurations from cited reports. First-party reported result; comparator results retain the source evaluation setup. MAI-Thinking-1 technical report · Table 11, PDF page 53 · reviewed 2026-09-06 | MAI-Thinking-1 technical report | lab self-report | 2026-09-06 |
| MAI-Thinking-1Microsoft | 84.2%Reported settings & sourceMAI avg 4 runs, temperature 1, top-p .97; simple ReAct bash/string-replace; Terminal-Bench ignores predefined timeouts. Comparator configurations from cited reports. First-party reported result; comparator results retain the source evaluation setup. MAI-Thinking-1 technical report · Table 11, PDF page 53 · reviewed 2026-09-06 | MAI-Thinking-1 technical report | lab self-report | 2026-09-06 |