Skip to content
ModelCap

Mistral Small 3

All-versions archive #149Arena #277Open weights

Mistral AI·mistralai/mistral-small-24b-instruct-2501

Mistral Small 3 is a language model published by Mistral AI. 1 OpenRouter endpoint is published in the latest snapshot. It accepts text input and returns text. The published context limit is 33K tokens. The catalogue declares support for structured outputs.

ModelCap Index · All-versions archive
32.7#149
Measured · 81% support · score interval 32.7–39.7
Market Gravity
34.7
Secondary market signal

Superseded by Mistral Small 4. It no longer holds a current-board position; its all-versions archive rank is #149.

Input
$0.05
per 1M tokens
Output
$0.08
per 1M tokens
Context
33K
tokens
Providers
1
endpoint
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
DeepInfrafp8$0.05$0.0833Kfp8100.0%

Reported evaluation leads (1)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
TIGER-Lab/MMLU-Pro
mmlu_pro
—66.3
Unpinned
not published
Reported
Quarantined

Other versions (1)

ModelCap Index · All-versions archive

Methodology
32.7
Measured · 2 public benchmark observations across 2 boards
All-versions archive
#149
Public rank basis
Measured (measured)
Index support
81.1% · strong
Index interval
32.7–39.7
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Identity binding
Exact catalogue product · mistralai/mistralai/mistral-small-24b-instruct-2501 · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata mistralai/Mistral-Small-24B-Instruct-2501@9527884be6e5616bdd54de542f9ae13384489724 (not the evaluation revision)
Direct benchmark observations
2
Breadth-adjusted BES point estimate
36.2
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
36.2
Evidence status
confirmed · 82% mass
General preference
35.8
Coding
37.6
Agents & tools
—
Reasoning
—
Evidence breadth
92%
Evidence coverage
81%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
12.6
Liquidity (providers × uptime)
77.8
Open reach (HF downloads)
51.6
Surface (context / tools / modalities)
15.1
Freshness
12.1
  • · OpenRouter popularity #145
  • · 1 live provider
  • · HF 30d downloads 68591
  • · Catalogue: context 32768
  • · First seen 2025-01-30

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index
82.2
strong · 98% metadata coverage
Not capability
Artifact reproducibility
100.0
Access & legal clarity
100.0
Deployability
74.0
Evaluation provenance
50.1

Missing: evaluationProvenance.evaluation-dataset-revisions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
34.7
OpenRouter weekly popularity
#145 of 298

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

Status
Ranked
Source
Arena
Official rank
#277 of 385
Rating
1274.2
95% confidence interval
1268–1280
Votes
15K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
99%

Market Gravity breakdown

Usage55%
12.6

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
77.8

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
51.6

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
12.1

Time since first public availability, on a six-month half-life.

Specification

Context window
33K tokens
Max output
16K tokens
Inputs
Text
Outputs
Text
Tokenizer
Mistral
Cached input tokens
Not offered
First seen on OpenRouter
30 Jan 2025
Knowledge cutoff
2023-10-31

Capabilities

  • Not supported: Reasoning
  • Not supported: Tool use
  • Supported: Structured output
  • Supported: Response format
  • Not supported: Moderated

Weights & access

Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.

Access
Open weights
Licence
apache-2.0
Pinned revision
9527884be6e5
Parameters
23.6B
Architecture
MistralForCausalLM
Model type
mistral
Hugging Face downloads (30d)
69K
Downloads (all time)
7.2M
Likes
964
Repository
mistralai/Mistral-Small-24B-Instruct-2501
Repository updated
28 Jul 2025, 17:26 UTC
Base lineage
mistralai/Mistral-Small-24B-Base-2501