Mistral
Mistral Large
mistral-largeContext
262K
Input / 1M tokens
$0.5000
Output / 1M tokens
$1.50
Weighted tokens formula
1× provider × 1× tier
About this model
Mistral Large
Mistral Large — Mistral AI's flagship dense model. Strong general-purpose quality with particularly good multilingual performance for European languages.
Released
2024-07
Training cutoff
2024-07
Parameters
123B (dense)
Best for
Use cases
- ✓European-language customer support
- ✓GDPR-resident workloads (data processed in EU)
- ✓Code generation
- ✓Structured output + function calling
Strengths
What it does well
- ●EU data residency by default
- ●Native function calling + JSON mode
- ●Strong on French, German, Spanish, Italian benchmarks
- ●Code quality matches GPT-4-class on common tasks
Limitations
Know the trade-offs
- !Trails frontier on the hardest reasoning tasks
- !Knowledge cutoff trails newer models
- !Premium price relative to open-weights at similar quality
Benchmarks
Published scores
Scores from official model cards and public leaderboards. Higher is better unless noted.
| Benchmark | Score | Measures |
|---|---|---|
| MMLU | 84.0 | General knowledge across 57 subjects |
| HumanEval | 92.0 | Python code generation, pass@1 |
| MATH | 60.4 | Mathematics, mixed difficulty |
Code example
Chat completion with Mistral Large
from openai import OpenAI
client = OpenAI(
base_url="https://api.anoman.io/v1",
api_key="anm-sk-..."
)
response = client.chat.completions.create(
model="mistral-large",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)Weighted tokens
weighted_tokens = raw_tokens × 1 (provider) × 1 (tier)
Pro plan: 20M weighted tokens/month. Combined multiplier 1×: 20,000,000 raw tokens available.
Alternatives
Similar models in our catalog
Workloads that fit Mistral Large often fit these too. Compare benchmarks and pricing before locking in a default.