{"schema":"postcutoff/event@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/e/2026-07-02-mistral-leanstral-1-5/","md":"https://postcutoff.com/e/2026-07-02-mistral-leanstral-1-5/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-07-02-mistral-leanstral-1-5","date":"2026-07-02","date_precision":"day","short_title":"Mistral releases Leanstral 1.5, an open-weights Lean 4 proof agent that saturates miniF2F and solves 587/672 PutnamBench problems","deck":null,"takeaway":"On July 2, 2026 Mistral released Leanstral 1.5 (\"Proof abundance for all\"), an Apache-2.0, 119B-parameter (6B active) MoE model specialised for Lean 4 theorem proving and code verification.","category":"open-source","category_label":"Open source","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"Mistral AI: Leanstral 1.5, proof abundance for all","url":"https://mistral.ai/news/leanstral-1-5","type":"official","group":"primary","domain":"mistral.ai"},{"n":2,"title":"Hugging Face: mistralai/Leanstral-1.5-119B-A6B","url":"https://huggingface.co/mistralai/Leanstral-1.5-119B-A6B","type":"code","group":"primary","domain":"huggingface.co"},{"n":3,"title":"Mistral AI for Developers on X: Leanstral release (March 16, 2026)","url":"https://x.com/MistralDevs/status/2033621477736477077","type":"official","group":"primary","domain":"x.com"},{"n":4,"title":"NYU Shanghai RITS: Leanstral, Mistral's open-source proof agent for Lean 4","url":"https://rits.shanghai.nyu.edu/ai/leanstral-mistrals-open-source-proof-agent-for-lean-4/","type":"press","group":"press","domain":"rits.shanghai.nyu.edu"}],"official":3,"filed":"2026-10-01","updated":"2026-10-01","orgs":["Mistral AI"],"title":"Mistral releases Leanstral 1.5, an open-weights Lean 4 proof agent that saturates miniF2F and solves 587/672 PutnamBench problems","summary":"On July 2, 2026 Mistral released Leanstral 1.5 (\"Proof abundance for all\"), an Apache-2.0, 119B-parameter (6B active) MoE model specialised for Lean 4 theorem proving and code verification. It updates the original Leanstral of March 16, 2026. Mistral reports 100% on miniF2F, 587 of 672 PutnamBench problems and state-of-the-art FATE-H/FATE-X scores. The weights are on Hugging Face, and a free API endpoint (`leanstral-1-5`) puts strong formal-proof tooling in anyone's hands.","key_facts":["Released July 2, 2026; 119B total / ~6B active parameters (MoE, 128 experts, 4 active per token); Apache 2.0; 256K context (≤200K recommended)","Benchmarks (Mistral): miniF2F 100% (validation and test); PutnamBench 587/672; FATE-H 87%; FATE-X 34%; FLTEval pass@1 28.9%, pass@8 43.2%","Test-time scaling on PutnamBench: 44 problems solved at a 50k-token budget, rising monotonically to 587 at 4M tokens","Training: mid-training, SFT, then RL with CISPO in two environments (multi-turn theorem proving and code-agent workflows)","Code verification: found 5 previously unreported bugs across 57 open-source repositories, including an integer overflow in datrs/varinteger's zigzag decoding","Predecessor: Leanstral (Leanstral-2603), released March 16, 2026 as 'the first open-source code agent for Lean 4' (Mistral)","Access: Hugging Face mistralai/Leanstral-1.5-119B-A6B; free Mistral API endpoint, model id leanstral-1-5"],"key_numbers":[],"tags":["lean","formal-verification","theorem-proving","open-weights","mistral","math","moe"],"science":null,"body_md":"## What happened\n\nMistral published Leanstral 1.5 on July 2, 2026. It is a sparse MoE model trained to work inside real Lean 4 repositories, writing proofs as well as\nformal specifications and proofs about software. The release is an update of Leanstral (March 2026), which Mistral had billed as the first open-source\ncode agent for Lean 4. The headline numbers are a saturated miniF2F (100%) and 587/672 on PutnamBench, with performance that keeps rising as the\ntoken budget grows to 4M tokens. Mistral also ran it as a code-verification agent over 57 open-source repositories and reports 5 new bugs.\n\n## Why it matters\n\nFormal proof was a frontier-lab capability (AlphaProof, Aristotle, Seed-Prover) through 2025. By mid-2026 an Apache-licensed model with only ~6B\nactive parameters reports near-saturation of the standard competition benchmarks, and it is free to call. That makes Lean verification cheap\nenough to bolt onto the summer-2026 wave of AI-generated mathematics and code. Benchmark numbers are Mistral's own and were not independently checked here.","disputed":[],"related":[{"id":"2026-09-21-lean-pool-ai-maintained-formal-math-archive","url":"https://postcutoff.com/e/2026-09-21-lean-pool-ai-maintained-formal-math-archive/","date":"2026-09-21","date_precision":"day","short_title":"Lean Pool: an AI-maintained archive of Lean formalizations grows past 3 million lines","deck":null,"takeaway":"It is an early example of mathematical infrastructure run mostly by AI agents, with humans as maintainers and contributors.","category":"research","category_label":"Research","importance":2,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":3,"official":3,"filed":"2026-09-30","updated":"2026-09-30","orgs":["Lean Pool"]},{"id":"2026-08-18-palomar-lean-registry","url":"https://postcutoff.com/e/2026-08-18-palomar-lean-registry/","date":"2026-08-18","date_precision":"day","short_title":"Palomar launches: a registry of Lean-verified mathematics to curb misrepresented AI proof claims","deck":null,"takeaway":"Formal verification became the main way to trust AI mathematics in 2026.","category":"science","category_label":"Science & math","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":6,"official":6,"filed":"2026-09-29","updated":"2026-10-29","orgs":["Lean FRO","ICARM"]},{"id":"2025-12-06-axiomprover-putnam-2025","url":"https://postcutoff.com/e/2025-12-06-axiomprover-putnam-2025/","date":"2025-12-06","date_precision":"day","short_title":"AxiomProver produces machine-checked Lean proofs for all 12 Putnam 2025 problems","deck":null,"takeaway":"Formal verification removes grading disputes like those around informal IMO proofs, and showed formal provers catching up with informal LLMs on hard competition maths.","category":"science","category_label":"Science & math","importance":3,"confidence":"medium","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":2,"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Axiom Math"]}],"people":[],"posts":[{"id":"x-mistraldevs-2033621477736477077","title":"“🧮 Today, we release Leanstral - the first open-source code agent for Lean 4, an…”","url":"https://postcutoff.com/p/x-mistraldevs-2033621477736477077/"}],"videos":[],"models":[{"id":"leanstral-1-5","name":"Leanstral 1.5","url":"https://postcutoff.com/m/leanstral-1-5/"}],"changes":[{"date":"2026-10-01","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-01","run":null,"sources_read":null,"updated":"2026-10-01","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":32,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":2,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":93,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":32,"in_training_data":false}],"short_url":null}