{"slug": "deepseek-math-v2", "name": "DeepSeek-Math-V2", "category": "ai-models", "category_name": "AI models", "description": "", "recommenders": 3, "commenters": 4, "warners": 1, "score": 1.815, "rising": -0.98, "first": "2025-11-27", "latest": "2026-04-25", "why": "if you want to do math proofs use AI built for proof", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 0, "compare": 0, "warn": 1}, "spark": [0, 0, 1, 1, 0, 2, 0, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "if you want to do math proofs use AI built for proof", "kind": "", "icon": true, "domain": null, "github_repo": "github.com/deepseek-ai/deepseek-math-v2", "official_url": "https://github.com/deepseek-ai/DeepSeek-Math-V2", "pricing": "", "aliases": ["DeepSeek-Math-V2", "DeepSeekMath-V2"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 766407584, "item_id": 47900474, "url": "https://news.ycombinator.com/item?id=47900474", "by": "simonjgreen", "date": "2026-04-25", "version": "2", "stance": "recommend", "quote": "your point is super valid and I agree with you!", "excerpt": "You are of course specifically referring to the math optimised models, not the chat ones folks would generally encounter. Not that I\u2019m trying to contradict you, your point is super valid and I agree with you! But I\u2019m supplementing to help anyone following along who may make choices. This is when it happened for anyone interested: https://binaryverseai.com/deepseek-math-v2-benchmarks-review-gemini-use/", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeek v4"}, {"id": 766234896, "item_id": 47889681, "url": "https://news.ycombinator.com/item?id=47889681", "by": "segmondy", "date": "2026-04-24", "version": "2", "stance": "recommend", "quote": "They have had the best math models for about a year", "excerpt": "They have had the best math models for about a year most folks just didn't know about it. You can't find inference on APIs, but I run these at home, this is also the advantage of open models. https://huggingface.co/deepseek-ai/DeepSeek-Math-V2 https://huggingface.co/deepseek-ai/DeepSeek-Prover-V2-671B", "context": "", "use_case": "Math problem solving", "switched_from": "", "story": "DeepSeek v4"}, {"id": 752141840, "item_id": 47008865, "url": "https://news.ycombinator.com/item?id=47008865", "by": "alansaber", "date": "2026-02-13", "version": "2", "stance": "recommend", "quote": "Deepseek wrote a decent paper on this", "excerpt": "Deepseek wrote a decent paper on this https://github.com/deepseek-ai/DeepSeek-Math-V2/blob/main/DeepSeekMath_V2.pdf", "context": "", "use_case": "", "switched_from": "", "story": "GPT-5.2 derives a new result in theoretical physics"}, {"id": 748181376, "item_id": 46761336, "url": "https://news.ycombinator.com/item?id=46761336", "by": "segmondy", "date": "2026-01-25", "version": "2", "stance": "recommend", "quote": "if you want to do math proofs use AI built for proof", "excerpt": "if you want to do math proofs use AI built for proof https://huggingface.co/deepseek-ai/DeepSeek-Math-V2 https://huggingface.co/deepseek-ai/DeepSeek-Prover-V2-671B", "context": "", "use_case": "math proofs", "switched_from": "", "story": "Case study: Creative math \u2013 How AI fakes proofs"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 737222944, "item_id": 46076434, "url": "https://news.ycombinator.com/item?id=46076434", "by": "hooloovoo_zoo", "date": "2025-11-27", "version": "2", "stance": "warn", "quote": "it's not a real score; they judged the results themselves", "excerpt": "For one thing, it's not a real score; they judged the results themselves and Putnam judges are notoriously tough. There was not a single 8 on the problem they claim partial credit for (or any partial credit above a 2) amongst the top 500 humans. https://kskedlaya.org/putnam-archive/putnam2024stats.html. For another thing, the 2024 Putnam problems are in their RL data. Also, it's very unclear how these competitions co", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning [pdf]"}]}], "who": [], "tags": [], "use_cases": ["math problem solving", "math proofs"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=47884971", "title": "DeepSeek v4", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=47006594", "title": "GPT-5.2 derives a new result in theoretical physics", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46759352", "title": "Case study: Creative math \u2013 How AI fakes proofs", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46072786", "title": "DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning [pdf]", "mentions": 1}], "sentiment_by_month": {"2025-11": {"warn": 1}, "2025-12": {}, "2026-01": {"recommend": 1}, "2026-02": {"recommend": 1}, "2026-03": {}, "2026-04": {"recommend": 2}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "quotes": [{"id": 766407584, "item_id": 47900474, "url": "https://news.ycombinator.com/item?id=47900474", "by": "simonjgreen", "date": "2026-04-25", "version": "2", "stance": "recommend", "quote": "your point is super valid and I agree with you!", "excerpt": "You are of course specifically referring to the math optimised models, not the chat ones folks would generally encounter. Not that I\u2019m trying to contradict you, your point is super valid and I agree with you! But I\u2019m supplementing to help anyone following along who may make choices. This is when it happened for anyone interested: https://binaryverseai.com/deepseek-math-v2-benchmarks-review-gemini-use/", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeek v4"}, {"id": 766234896, "item_id": 47889681, "url": "https://news.ycombinator.com/item?id=47889681", "by": "segmondy", "date": "2026-04-24", "version": "2", "stance": "recommend", "quote": "They have had the best math models for about a year", "excerpt": "They have had the best math models for about a year most folks just didn't know about it. You can't find inference on APIs, but I run these at home, this is also the advantage of open models. https://huggingface.co/deepseek-ai/DeepSeek-Math-V2 https://huggingface.co/deepseek-ai/DeepSeek-Prover-V2-671B", "context": "", "use_case": "Math problem solving", "switched_from": "", "story": "DeepSeek v4"}, {"id": 752141840, "item_id": 47008865, "url": "https://news.ycombinator.com/item?id=47008865", "by": "alansaber", "date": "2026-02-13", "version": "2", "stance": "recommend", "quote": "Deepseek wrote a decent paper on this", "excerpt": "Deepseek wrote a decent paper on this https://github.com/deepseek-ai/DeepSeek-Math-V2/blob/main/DeepSeekMath_V2.pdf", "context": "", "use_case": "", "switched_from": "", "story": "GPT-5.2 derives a new result in theoretical physics"}, {"id": 748181376, "item_id": 46761336, "url": "https://news.ycombinator.com/item?id=46761336", "by": "segmondy", "date": "2026-01-25", "version": "2", "stance": "recommend", "quote": "if you want to do math proofs use AI built for proof", "excerpt": "if you want to do math proofs use AI built for proof https://huggingface.co/deepseek-ai/DeepSeek-Math-V2 https://huggingface.co/deepseek-ai/DeepSeek-Prover-V2-671B", "context": "", "use_case": "math proofs", "switched_from": "", "story": "Case study: Creative math \u2013 How AI fakes proofs"}, {"id": 737222944, "item_id": 46076434, "url": "https://news.ycombinator.com/item?id=46076434", "by": "hooloovoo_zoo", "date": "2025-11-27", "version": "2", "stance": "warn", "quote": "it's not a real score; they judged the results themselves", "excerpt": "For one thing, it's not a real score; they judged the results themselves and Putnam judges are notoriously tough. There was not a single 8 on the problem they claim partial credit for (or any partial credit above a 2) amongst the top 500 humans. https://kskedlaya.org/putnam-archive/putnam2024stats.html. For another thing, the 2024 Putnam problems are in their RL data. Also, it's very unclear how these competitions co", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning [pdf]"}]}