ModelPriceWatch.com
Last scan 2026-08-01 Models tracked 181 Providers 27 Cheapest paid Granite 4.0 Micro $0.017/Mtok in Every price links to its source

Mistral Small 3.2 24B

by Mistral · 24B parameters

Retired budget open weights cheap tier
Retired on Jul 31, 2026. No longer served on the endpoint we price — the figures below are kept for historical reference only. Replaced by Mistral Small 4.
Availability: Retired 2026-07-31. mistral-small-2506 now appears in Mistral’s “Deprecated & retired models” table (deprecated 2026-04-30, retired 2026-07-31); the API endpoint is withdrawn. Use Mistral Small 4. The price shown is the last list price we recorded while it was served; open weights remain downloadable.

Today's price · per 1M tokens

Input

$0.100

per 1M tokens

Output

$0.300

per 1M tokens

Blended

$0.200

avg of input & output $/1M

Source: official Mistral pricing · read Aug 1, 2026 MODELPRICEWATCH.COM · 2026-08-01

Overview

Open-weights small model. $0.10/$0.30 per 1M tokens. Retired from the Mistral API 2026-07-31.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 1/5
Text ✓ Image Audio Video PDF
Output 1/5
Text ✓ Image Audio Video Embedding
Features 3/9
Prompt caching Reasoning Coding Fast inference ✓ Long context ✓ Open weights ✓ Multimodal Web search Realtime

Benchmark performance

accuracy % · higher is better
Percentile vs all tracked models 10.6th1 independent measurement
Percentile per $/Mtok 53
GPQA Diamond vendor
46.13%

Every per-benchmark score we hold for this model is reported by its own vendor, so none is counted toward an accuracy average or any ranking. The percentile above is its standing across the independent composites below.

Independent composite scores
  • AA Intelligence Index: 11 (Artificial Analysis composite across reasoning, knowledge and coding evals)
How it stacks up
  • Ranks #113 of 133 comparably-measured models by percentile score, across 1 independent measurement
  • Ranks #33 of 133 comparably-tested models by normalized performance per dollar

148 tokens/sec output

Source: Artificial Analysis, Mistral model card · updated Jul 18, 2026 · See full rankings →

Specifications

Provider
Mistral
Context window
128K tokens
Modality
text
Parameters
24B
Open source
Yes — open weights available
Released
Jun 1, 2025
Status
Retired ended Jul 31, 2026
Last updated
Aug 1, 2026
Tags
budgetopen-weightsfast

Availability verified: Aug 1, 2026 — per Mistral's own deprecation notice