Voxtral Small 24B
Not tested yet
Language modelActive
Voxtral Small 24B
Mistral AI
- Released
- July 15, 2025
- Data date
- September 8, 2026
- Intelligence
- Overall score
- Price
- Input · Output / 1M tokens
- Input
- Text, Audio
- Output
- Text
No published category-wide ranking is available for this model.
Entries 1-12 of 51. Page 1 of 5.
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
Speech Understanding LLM judge (Show measurement, test conditions, and source)
- Source value
- 86.61
- Score
- 86.61
- Metric
- percent
- Unit
- %
- Category
- speech-understanding
- Direction
- Higher is better
- Comparison cohort
- research:voxtral-small-24b:0
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 2
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); SU benchmark; mean over ten independent judge trials
English Short-Form WER (macro) (Show measurement, test conditions, and source)
- Source value
- 6.31
- Score
- 6.31
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:2
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 2
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); macro-average WER across English short-form test sets
LibriSpeech Test Clean WER (Show measurement, test conditions, and source)
- Source value
- 1.53
- Score
- 1.53
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:3
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; LS-C
LibriSpeech Test Other WER (Show measurement, test conditions, and source)
- Source value
- 3.14
- Score
- 3.14
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:4
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; LS-O
GigaSpeech WER (Show measurement, test conditions, and source)
- Source value
- 10.27
- Score
- 10.27
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:5
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; GS
VoxPopuli WER (Show measurement, test conditions, and source)
- Source value
- 6.62
- Score
- 6.62
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:6
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; VP
SwitchBoard WER (Show measurement, test conditions, and source)
- Source value
- 11.09
- Score
- 11.09
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:7
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; SB
CHiME-4 WER (Show measurement, test conditions, and source)
- Source value
- 9.64
- Score
- 9.64
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:8
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; C-4
SPGISpeech WER (Show measurement, test conditions, and source)
- Source value
- 1.89
- Score
- 1.89
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:9
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); short-form English speech recognition; SPGI
Earnings-21 10m WER (Show measurement, test conditions, and source)
- Source value
- 9.55
- Score
- 9.55
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:10
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); long-form English speech recognition; E21 10m
Earnings-22 10m WER (Show measurement, test conditions, and source)
- Source value
- 12.48
- Score
- 12.48
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:11
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 3
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); long-form English speech recognition; E22 10m
FLEURS WER (Arabic) (Show measurement, test conditions, and source)
- Source value
- 13.44
- Score
- 13.44
- Metric
- percent
- Unit
- %
- Category
- speech-recognition
- Direction
- Lower is better
- Comparison cohort
- research:voxtral-small-24b:12
- Source type
- vendor-reported
- Evaluator
- Voxtral technical report Table 4
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. Voxtral Small SFT (public checkpoint); per-language FLEURS speech recognition; ar
Our model tests
Community tests you can inspect
Compare how the models respond to the same prompt. Each test shows the first attempt, with no subsequent fixes to the generated code. These results do not contribute to an overall score.
Pelican on a bicycle
A community classic for free-form SVG drawing.
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)Profile
Specifications and access
Published information about this model. Existing estimates are explicitly labeled.
Additional source observations
Dated documentation and model-card observations. Provider limits, native context and extended context can differ. Configuration notes retain the source wording.
- License
- Apache-2.0Source
- Release
- 2025-07-15 (Inventory lists July 1; official release post is used.)Source
- Parameters
- 24 BSource
- Architecture
- audio-language model using Mistral Small 3.1 text backboneSource
- Context window
- 32,768 tokensSource
- Input
- Audio, TextSource
- Output
- TextSource
- Audio limits
- up to 30 minutes transcription, up to 40 minutes audio understandingSource
- Languages
- English, French, German, Italian, Spanish, Portuguese, Dutch, HindiSource
- Tool calling
- YesSource
Pricing
Published prices
Prices apply to the stated unit. Resolution, output length, and provider can change the cost.
Cost example
100 API requests
$0.025
100 requests with 1,000 input and 500 output tokens each calculate to $0.025. Input accounts for $0.01; output accounts for $0.015.
- Input price
- $0.1 / 1M tokensBase tier for each request
- Output price
- $0.3 / 1M tokensBase tier for each request
The calculation uses documented token prices and no cache discount. It excludes extra tools, tax, reasoning tokens, and further hidden output tokens.
Open sourceHead-to-head comparisons
Compare this model
Each matchup compares this model with exactly one other model from the same category.
There are no published direct comparisons for this model yet.
More models
Models from the same selection
All AI models
Evidence
Sources and data date
Every statement links to its underlying documentation or leaderboard.
- Price
Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs.
- Input
Documented modalities. Retrieved September 8, 2026.
- Output
Documented modalities. Retrieved September 8, 2026.
Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Open evidence questions and research notes (original wording)
Not found in the inspected sources
knowledge-cutoff. No exact cutoff found.
Not found in the inspected sources
training-tokens. No token count found.
Not found in the inspected sources
max-output-tokens. No model-specific maximum output found.
Unresolved
AA benchmark rows. No exact Artificial Analysis model page located in the checked public sitemap.