Requesty
Mistral AI SAS logo
Mistral AI SAS

leanstral-1-5

Leanstral 1.5 is an updated Lean 4 formal proof engineering model from Mistral AI, optimized for automated theorem proving and autoformalization. It has 119B total parameters with 6.5B active and supports a 256K token context window. It supports native function calling and structured output.

Tool callingJSON schema
Try leanstral-1-5CompareDocs

Specifications

Context window262K tokens
Max output33K tokens
API typechat
AddedJul 3, 2026
Model IDmistral/leanstral-1-5
Data retentionYes (30 days)
Used for trainingNo
Provider location🇪🇺 EU
Privacy policyMistral Privacy Policy→

Benchmarks

Released 2023-12-11
GPQA Diamondreasoning
34.9%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
3.6%

Artificial Analysis Intelligence Index — a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality. Always test on your own workload.

Pricing

Input / 1M
Free
Output / 1M
Free
Cache write
N/A
Cache read
N/A
Estimated cost
100K input + 10K outputFree
1M input + 100K outputFree
10M input + 1M outputFree

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to mistral/leanstral-1-5.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="mistral/leanstral-1-5",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Mistral AI SAS models

codestral-latest
131K ctx$0.30/1M
devstral-latest
256K ctx$0.40/1M
open-mistral-7b
33K ctx$0.25/1M
mistral-medium-latest
131K ctx$0.40/1M
mistral-small-2503
33K ctx$0.10/1M
devstral-small-2507
131K ctx$0.10/1M

Frequently asked questions

How much does leanstral-1-5 cost?
leanstral-1-5 is priced at Free per million input tokens and Free per million output tokens when accessed via Requesty. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of leanstral-1-5?
leanstral-1-5 has a context window of 262K tokens, with a maximum output of 33K tokens per response. That's roughly 350 words of input you can fit in a single prompt.
How does leanstral-1-5 perform on benchmarks?
leanstral-1-5 scores 49.1% on MMLU Pro, 34.9% on GPQA Diamond, 11.8% on SciCode. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can leanstral-1-5 do?
leanstral-1-5 supports tool calling, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use leanstral-1-5 with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "mistral/leanstral-1-5". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run leanstral-1-5 through Requesty?
Yes. leanstral-1-5 runs through Requesty's OpenAI-compatible API, served from Mistral AI SAS. You do not host the model yourself: point base_url at Requesty, set the model to "mistral/leanstral-1-5", and requests are routed to the upstream provider with automatic failover. The same key gives you 400+ other models too.

Access leanstral-1-5 through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Mistral AI SAS models