Mistral Small 3.2 24B
Mistral · Open weights · Released Jun 20, 2025 · mistralai/mistral-small-3.2-24b-instruct
Best $0.075 in / $0.20 out via DeepInfra · fp8 · 128K ctx at this price (headline 256K)
256K contextVisionToolsStructured outputPrompt cachingOpen weights
Available from 4 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 17 MINUTES AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
| $0.075 | $0.20 | — | 128K | fp8 | ✕ | ✕ | 100.0% | |
| $0.09 | $0.30 | $0.05 | 131K | bf16 | 1.5s | 64 tok/s | 100.0% | |
| $0.094 | $0.25 | — | 256K | fp8 | ✕ | ✕ | 99.9% | |
| $0.10 | $0.30 | $0.01 | 33K | — | ✕ | ✕ | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · methodology v1.1 · measured 17 minutes ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Recent changesRSS ↗
Sep 25, 2026endpoint added · mistral3 → 4
Sep 12, 2026endpoint added · venice2 → 3
Sep 12, 2026context changed131072 → 256000
Aug 31, 2026endpoint removed · mistral3 → 2
Aug 24, 2026endpoint added · mistral2 → 3
About
Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on...
Specifications
Context window256,000
Max output16,384
Modalities inimage, text
Modalities outtext
LicenseOpen weights
ReleasedJun 20, 2025
Knowledge cutoff2023-10-31
TokenizerMistral
WeightsHugging Face ↗
Open weights
safetensors
Related
Embed badge
[](https://modelindex.ai/models/mistralai/mistral-small-3.2-24b-instruct)Sources
OpenRouter API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks17 minutes ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.