Voxtral Small 24B 2507
Mistral AIhuggingface.co/mistralai/Voxtral-Small-24B-2507
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...
Updated 31 min ago · first seen 11 Sept 2026
model_01M294WW9GHMY2KK38BWM7A7YS
- Parameters
- 24.3B
- T2 · 34 min ago
- Context
- 32.8K tokens
- T2 · 35 min ago
- Released
- 30 Oct 2025
- T2 · 35 min ago
Specification
- Release date
- 30 Oct 2025
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 34 min agomedium
- Parameters
- 24.3B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 34 min agomedium
- Context window
- 32.8K tokens
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Max output
- 26.2K tokens
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Modalities
- audio, text
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Input modalities
- audio, text
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Tokenizer
- Mistral
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Hugging Face repo
- mistralai/Voxtral-Small-24B-2507
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Pipeline tag
- audio-text-to-text
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 34 min agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 34 min agomedium
- Downloads
- 127,041
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 31 min agomedium
- Likes
- 524
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 31 min agomedium
- File input
- Yes
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Last modified
- 2025-12-20T23:34:49+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 34 min agomedium
- Openrouter id
- mistralai/voxtral-small-24b-2507
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Supported parameters
- frequency_penalty, max_tokens, presence_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
21
Source tiers
T221
Freshest observation
31 min ago
Conflicts
None
Modalities
- Modalities
- audiotext
- Input
- audiotext
- Output
- text
Capabilities
Tool calling
Yes
OpenRouter public model & pricing listing · T2
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Unavailable
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 32.8K tokens
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Max output
- 26.2K tokens
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
- Tokenizer
- Mistral
Source:OpenRouter public model & pricing listingT2observed 35 min agomedium
No benchmark results recorded
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| Mistral AI La Plateformemistralai/voxtral-small-24b-2507 | $0.10 | $0.30 | $0.01 | — | 32.8K | 35 min ago | openrouter.aiT2 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 1
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- Mistral AI La Plateforme
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| Apple M3 Ultra | 4bit | — | 14.4 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 14.4 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 14.4 GB | Yes |
| Apple M2 Ultra | 4bit | — | 14.4 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 14.4 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 14.4 GB | Yes |
| Apple M1 Ultra | 4bit | — | 14.4 GB | Yes |
| Apple M3 Max | 4bit | — | 14.4 GB | Yes |
| Apple M4 Max | 4bit | — | 14.4 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 14.4 GB | Yes |
| Apple M2 Max | 4bit | — | 14.4 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 14.4 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 14.4 GB | Yes |
| Apple M1 Max | 4bit | — | 14.4 GB | Yes |
| Apple M4 Pro | 4bit | — | 14.4 GB | Yes |
| Apple M3 Pro | 4bit | — | 14.4 GB | Yes |
| Apple M1 Pro | 4bit | — | 14.4 GB | Yes |
| Apple M2 Pro | 4bit | — | 14.4 GB | Yes |
| Apple M4 | 4bit | — | 14.4 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 14.4 GB | Yes |
| Apple M2 | 4bit | — | 14.4 GB | Yes |
| Apple M3 | 4bit | — | 14.4 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 14.4 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 14.4 GB | Yes |
| Apple M1 | 4bit | — | 14.4 GB | No |
No lineage recorded
Papers 0
No papers linked yet.
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Supported parameters
Supported parameterssupported_parameters1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| frequency_penalty, max_tokens, presence_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p | → current | current | OpenRouter public model & pricing listingT2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
New model: Voxtral Small 24B 2507 (Mistral AI)
openrouterMistral AI La Plateforme lists Voxtral Small 24B 2507 at $0.1 in / $0.3 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=mistralai&p=0&sort=downloads | listing | T2· Quality secondary | 31 min ago | 2 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 35 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.