Home/Models/Voxtral Small 24B 2507
Mistral AI logo

Voxtral Small 24B 2507

by Mistral AI · Released Oct 2025

Open SourceMultimodal
26.3
avg score
Rank #191
Compare
Better than 18% of all models
Context
32K tokens (~16 books)
Input $/1M
$0.10
Output $/1M
$0.30
Type
multimodal
License
Open Source
Benchmarks
1 tested
Data updated today
About

Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio...

Tested on 1 benchmarks with 26.3% average. Top scores: AudioMultiChallenge — Text Output (26.3%).

Capabilities
knowledge
26.3
#185 globally
Benchmark Scores
Compare All
Tested on 1 benchmarks · Ranked across 1 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
AudioMultiChallenge — Text Output

SEAL Audio Multi-Challenge (text output). Tests ability to describe and analyze audio input.

26.3
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
Community
BenchGecko API
voxtral-small-24b-2507
Specifications
  • Typemultimodal
  • Context32K tokens (~16 books)
  • ReleasedOct 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.001
Available On
Mistral AI logoMistral AI$0.10
Categories
Share & Export
Tweet
Voxtral Small 24B 2507 is an open-source multimodal AI model by Mistral AI, released in October 2025. It has an average benchmark score of 26.3. Context window: 32K tokens.