Data as of October 25, 2026 · How the score is built
Released Jul 1, 2026256K contextMistral Leanstral 1.5 model card
Leanstral 1.5
Decision readingLeanstral 1.5 is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Jul 1, 2026 — see all recent releases
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
LeanstralSpec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- leanstral-1-5Mistral Leanstral 1.5 model card
- Context window
- 256KMistral Leanstral 1.5 model card
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Mistral publishes the 119B-total, 6.5B-active Lean 4 proof-engineering agent under Apache 2.0 on Hugging Face (256K context, 200K recommended) and serves it as leanstral-1-5 on La Plateforme; the original Leanstral is scheduled for retirement on September 30, 2026.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordMistral Leanstral 1.5 model card
- Self-host
- Leanstral 1.5 needs ~80GB VRAM (1× NVIDIA H100 (80GB)).
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Leanstral 1.5, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
Leanstral 1.5 is a open weight model with a 256K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Mistral publishes the 119B-total, 6.5B-active Lean 4 proof-engineering agent under Apache 2.0 on Hugging Face (256K context, 200K recommended) and serves it as leanstral-1-5 on La Plateforme; the original Leanstral is scheduled for retirement on September 30, 2026.
Mistral’s Leanstral 1.5 model card (July 1, 2026) documents the updated Lean 4 formal-proof agent without a benchmark table, so BenchLM stores no scored rows. The row stays unranked.
Leanstral 1.5 sits in the Leanstral family with Leanstral. Its explicit predecessor is Leanstral. The profile has no source-displayable benchmark row yet.
Last updated October 25, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Questions
How does Leanstral 1.5 perform overall in AI benchmarks?
Leanstral 1.5 does not have any source-displayable benchmark rows yet, so this profile does not assign a public score or rank. Documented specifications remain visible, while score-led charts and claims stay unavailable until a published evaluation can be attached to the exact model.
Is Leanstral 1.5 open source?
Leanstral 1.5 is an open-weight model from Mistral. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to Leanstral 1.5?
Leanstral 1.5 belongs to the Leanstral family. Related tracked variants include Leanstral. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Leanstral 1.5 have full benchmark coverage on BenchLM?
No. Leanstral 1.5 currently has 0 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Leanstral 1.5?
Leanstral 1.5 has a documented context window of 256K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.