Mistral 7B Instruct
Intelligence
7.4
#503/523
Coding
—
Math
—
Speed
159 tok/s
TTFT: 285.00s
Pricing
$0.25 / $0.25
per 1M tokens (in/out)
Mistral 7B Instruct is Mistral’s offering for instruction-based tasks. It processes at 158.636 tokens per second and is priced at $0.25 per million tokens, targeting professional users.
When to Use Mistral 7B Instruct
✓ Best For
- Instruction-based applications
- Natural language understanding tasks
- Conversational agents
✗ Not Ideal For
- High-performance coding tasks
- Complex mathematical computations
How Mistral 7B Instruct Compares
Intelligence Index · Higher is better
AlibabaAnthropicMistralCohereAllen Institute for AI
Benchmark Profile
Output Speed · tok/s
NVIDIAOpenAIMistral
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (8)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 7.4 | |
| MMLU-Pro | 245% | |
| GPQA | 177% | |
| LiveCodeBench | 46% | |
| HLE | 43% | |
| SciCode | 2.4% | |
| IFBench | 19.9% | |
| MATH 500 | 12.1% |
Data: Artificial Analysis · Updated: April 2, 2026
Frequently Asked Questions (3)
What is Mistral 7B Instruct?
It is a language model designed for instruction-based tasks.
How much does Mistral 7B Instruct cost?
It costs $0.25 per million input tokens and $0.25 per million output tokens.
What is Mistral 7B Instruct best for?
It is best for instruction-based applications and natural language understanding tasks.
