{"slug": "darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open", "title": "Darwin-180B-RSI Tops MMLU-Pro and GPQA Diamond on AIhoster's Independent Open-Model Benchmark Board", "summary": "VIDRAFT's Darwin-180B-RSI, a 180-billion-parameter open-weight model, has taken the top spot on both MMLU-Pro (88.1%) and GPQA Diamond (94.4%) on AIhoster's independent benchmark aggregation board as of its 2026-10-19 update, edging out models with far larger parameter counts such as DeepSeek-V4-Pro and Kimi-K3. The GPQA figure was achieved with majority voting over up to 16 samples, while sibling VIDRAFT model Darwin-397B-ZTC reached 93.4% with greedy decoding. AIhoster aggregates publicly reported scores without recomputing them, and VIDRAFT has not disclosed architecture, training data, or licensing details.", "body_md": "**TL;DR:** VIDRAFT's Darwin-180B-RSI, a 180-billion-parameter open model, has claimed the #1 position on both the MMLU-Pro and GPQA Diamond leaderboards as tracked by Russian benchmark aggregator AIhoster — scoring 88.1% and 94.4% respectively, outperforming models with parameter counts an order of magnitude larger. Developers interested in frontier reasoning capabilities in an open-weight format should take note.\n\nDarwin-180B-RSI is a 180B-parameter open-weight language model developed by VIDRAFT (비드래프트), a Korean Pre-AGI AI startup. It is published under the Hugging Face repository identifier `FINAL-Bench/Darwin-180B-RSI`.\n\nAIhoster is an independent Russian benchmark aggregation site that collects publicly reported evaluation results from model cards, papers, and leaderboards — without recomputing or normalizing scores into a composite index. Each benchmark entry is tied to an exact repository ID and a specific evaluation configuration, making comparisons methodologically transparent.\n\nAs of the AIhoster update on 2026-10-19, Darwin-180B-RSI holds:\n\nNotably, VIDRAFT has multiple Darwin-family models appearing in the GPQA Diamond rankings simultaneously: Darwin-397B-ZTC (#3 at 93.4%) and Darwin-398B-JGOS (#9 at 90.9%), suggesting an active, iterative development pipeline across different model scales and inference configurations.\n\nAt a conceptual level, Darwin-180B-RSI appears to be a large-scale transformer-based language model optimized for extended reasoning. A few things can be inferred from the publicly disclosed evaluation configurations:\n\nNo architectural specifics, training hyperparameters, or dataset details are publicly disclosed in the source material.\n\nAll figures below are sourced directly from AIhoster's leaderboard (updated 2026-10-19), which in turn cites the Darwin-180B-RSI model card.\n\n**MMLU-Pro** (multi-domain expert knowledge; higher accuracy = better):\n\n| Rank | Model | Score | Params | \n|---|---|---|---|\n| 1 | **Darwin-180B-RSI** | **88.1%** | 180B | \n| 2 | MiniMax-M2.1 | 88.0% | 229B | \n| 3 | Intern-S2-Preview | 88.0% | 36.1B | \n| 4 | Qwen3.5-397B-A17B | 87.8% | 403B | \n| 5 | DeepSeek-V4-Pro | 87.5% | 1.6T | \n\n**GPQA Diamond** (expert-level physics, chemistry, biology; higher accuracy = better):\n\n| Rank | Model | Score | Params | \n|---|---|---|---|\n| 1 | **Darwin-180B-RSI** | **94.4%** | 180B | \n| 2 | Kimi-K3 | 93.5% | 2.8T | \n| 3 | Darwin-397B-ZTC | 93.4% | 404B | \n| 4 | Ornith-1.5-397B | 92.8% | 403B | \n| 5 | Qwen3.8-2.4T-A95B | 92.6% | 2.4T | \n\nThe GPQA result is particularly notable: Darwin-180B-RSI achieves 94.4% using majority voting over up to 16 samples, while the #3 model, Darwin-397B-ZTC (also a VIDRAFT model), achieves 93.4% with greedy decoding and no voting — a different inference regime entirely.\n\nThe model is listed under the Hugging Face organization `FINAL-Bench` with the repository `FINAL-Bench/Darwin-180B-RSI`. If the weights are publicly available there, you can browse or download via the standard Hugging Face CLI:\n\n```\nhuggingface-cli download FINAL-Bench/Darwin-180B-RSI\n```\n\nAccess details beyond what is on the model card — including any API endpoints, licensing terms, or inference service availability — are not confirmed in the source material. Check the model card directly for the most current access instructions.\n\n**Q: How does Darwin-180B-RSI beat models with 10× more parameters on MMLU-Pro?**\n\nA: The source doesn't explain internal architecture, but the evaluation configuration gives a clue: the model uses a 131,072-token thinking budget, suggesting extended chain-of-thought reasoning. Parameter efficiency at inference is a known benefit of well-optimized mixture-of-experts or distillation-based architectures, though VIDRAFT has not publicly confirmed which approach is used here.\n\n**Q: Are the GPQA and MMLU-Pro scores independently verified?**\n\nA: AIhoster explicitly states it does not recompute scores — it aggregates from public model cards and leaderboard sources, matched by exact repository ID. The source for Darwin-180B-RSI's scores is listed as its own model card. Independent third-party reproduction has not been confirmed in this reporting.\n\n**Q: Can I use Darwin-180B-RSI commercially?**\n\nA: Licensing terms are not specified in the source article. Check the model card at `FINAL-Bench/Darwin-180B-RSI` on Hugging Face for the applicable license before any production or commercial use.\n\n*Originally reported by AIhoster (러시아) (2026-10-18) — [source article](https://aihoster.ru/benchmarks).*", "url": "https://wpnews.pro/news/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open", "canonical_source": "https://dev.to/ai_openfree_b23025ef075cf/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihosters-independent-open-model-benchmark-board-17d7", "published_at": "2026-10-19 11:01:30+00:00", "updated_at": "2026-10-19 11:14:37.557640+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "artificial-intelligence"], "entities": ["VIDRAFT", "Darwin-180B-RSI", "AIhoster", "Hugging Face", "Darwin-397B-ZTC", "Kimi-K3", "DeepSeek-V4-Pro", "Qwen3.5-397B-A17B"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open", "markdown": "https://wpnews.pro/news/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open.md", "text": "https://wpnews.pro/news/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open.txt", "jsonld": "https://wpnews.pro/news/darwin-180b-rsi-tops-mmlu-pro-and-gpqa-diamond-on-aihoster-s-independent-open.jsonld"}}