Skip to content

Mistral Small 4

MistralOpen weightsNon-reasoningProvisionalReleased Mar 16, 2026Input: text, image
Compare with ZAYA1-8BSource scores ↗OpenRouter ↗
Overall score
34.7
Rank 161 of 222 · 25.0–44.4
Context window
256K
256,000 tokens
Price per 1M tokens
$0.15 in · $0.60 out
Source: BenchLM
Blended price
$0.26
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Reasoning59.1
  2. 2Instruction following#8557.1
  3. 3Multimodal48.2
  4. 4Knowledge44.1
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (0)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

No benchmark results with redistributable sources yet.

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (BenchLM)$0.15$0.60—
OpenRouter mistralai/mistral-small-2603$0.15$0.60$0.015

Mistral's official Mistral Small 4 model card lists $0.15 input / $0.60 output per million tokens and describes Mistral Small 4 as a single hybrid model for instruct, reasoning, and coding.

Open weights

From the Hugging Face repository.

Parameters
119B
Licence
apache-2.0
Downloads (30 days)
71.2K

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Similar score