Mistral Small 4 (Non-reasoning)
Intelligence
18.6
#277/523
Coding
16.4
#245/429
Math
—
Speed
157 tok/s
TTFT: 0.41s
Pricing
$0.15 / $0.60
per 1M tokens (in/out)
Mistral Small 4 (Non-reasoning) is Mistral’s model designed for efficient processing of text data. It operates at a speed of 156.729 tokens per second and is priced at $0.15 per million input tokens, targeting professional users.
When to Use Mistral Small 4 (Non-reasoning)
✓ Best For
- Text processing tasks
- Data analysis
- Basic coding assistance
✗ Not Ideal For
- Complex reasoning tasks
- High-level mathematical computations
How Mistral Small 4 (Non-reasoning) Compares
Intelligence Index · Higher is better
AnthropicMistralNous ResearchTrillion Labs
Benchmark Profile
Coding Index
LongCatAlibabaMistralDeepSeekMBZUAI Institute of Foundation Models
Output Speed · tok/s
OpenAIMistral
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (9)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 18.6 | |
| Coding Index | 16.4 | |
| GPQA | 571% | |
| HLE | 37% | |
| SciCode | 28.1% | |
| IFBench | 32.8% | |
| LCR | 21.3% | |
| TerminalBench Hard | 10.6% | |
| Tau2 | 18.4% |
Data: Artificial Analysis · Updated: April 9, 2026
Frequently Asked Questions (15)
When was Mistral Small 4 (Non-reasoning) released?
Mistral Small 4 (Non-reasoning) was released on March 16, 2026.
Who created Mistral Small 4 (Non-reasoning)?
Mistral Small 4 (Non-reasoning) was created by Mistral.
How intelligent is Mistral Small 4 (Non-reasoning)?
Mistral Small 4 (Non-reasoning) scores 19 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight non-reasoning models of similar size (median: 13).
How fast is Mistral Small 4 (Non-reasoning)?
Mistral Small 4 (Non-reasoning) generates output at 120.3 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 59.9 t/s).
What is the latency of Mistral Small 4 (Non-reasoning)?
Mistral Small 4 (Non-reasoning) has a time to first token (TTFT) of 1.34s (based on Mistral's API), which is very competitive compared to other open weight non-reasoning models of similar size (median: 1.90s).
How much does Mistral Small 4 (Non-reasoning) cost?
Mistral Small 4 (Non-reasoning) costs $0.15 per 1M input tokens (very competitive, median: $0.54) and $0.60 per 1M output tokens (better than average, median: $0.90), based on Mistral's API.
What is Mistral Small 4 (Non-reasoning) API pricing?
Mistral Small 4 (Non-reasoning) costs $0.15 per 1M input tokens and $0.60 per 1M output tokens (based on Mistral's API). For a blended rate (3:1 input to output ratio), this is $0.26 per 1M tokens. Pricing may vary by provider.
How verbose is Mistral Small 4 (Non-reasoning)?
When evaluated on the Intelligence Index, Mistral Small 4 (Non-reasoning) generated 3.9M output tokens, which is better than average compared to other open weight non-reasoning models of similar size (median: 6.4M).
Is Mistral Small 4 (Non-reasoning) a reasoning model?
No, Mistral Small 4 (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Mistral Small 4 (Non-reasoning) support?
Mistral Small 4 (Non-reasoning) supports text and image input.
What output modalities does Mistral Small 4 (Non-reasoning) support?
Mistral Small 4 (Non-reasoning) supports text output.
Can Mistral Small 4 (Non-reasoning) process images?
Yes, Mistral Small 4 (Non-reasoning) supports image input and can analyze, describe, and answer questions about images.
Is Mistral Small 4 (Non-reasoning) multimodal?
Yes, Mistral Small 4 (Non-reasoning) is multimodal. It can process text and image input and generate text output.
What is the context window of Mistral Small 4 (Non-reasoning)?
Mistral Small 4 (Non-reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Is Mistral Small 4 (Non-reasoning) open source?
Yes, Mistral Small 4 (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
