Mistral AI
European AI lab building efficient open-weight models
MultilingualStructured outputOCREfficiencyOpenAI-compatible
About Mistral AI
Mistral AI is a Paris-based company that builds high-performance language models with a focus on efficiency and openness. Their models range from the compact Mistral Small to the powerful Mistral Large, with strong multilingual capabilities and structured output support. Mistral also offers a specialized OCR API for document processing.
Key Models
Mistral LargeMistral SmallMistral NemoCodestral
Live Performance Data
Mistral AI Models on Antbase
| Model | Score | Requests | Success Rate | Avg Latency | p95 Latency | Capabilities |
|---|---|---|---|---|---|---|
| Mistral Vibe Cli Fast | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Medium Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Medium | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Vibe Cli With Tools | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Open Mistral Nemo | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Open Mistral Nemo 2407 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Mistral Tiny 2407 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Mistral Tiny Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Codestral 2508 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Codestral Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Devstral 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Mistral Vibe Cli Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Devstral Medium Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Pixtral Large 2411 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Pixtral Large Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Devstral Small 2507 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Devstral Medium 2507 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Labs Devstral Small 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Devstral Small Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Medium 2505 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Mini 2507 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Magistral Small 2509 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Ministral 3b 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Magistral Medium 2509 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Magistral Medium Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Magistral Small Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Small 2507 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Ministral 3b Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Small Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Ministral 8b 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Ministral 8b Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Ministral 14b 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Ministral 14b Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Mini Transcribe 2507 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Mistral Ocr 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Ocr Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Ocr 2503 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Ocr 2505 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Mini 2602 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Voxtral Mini Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Mistral Medium 2508 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Mini Realtime Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Devstral Latest | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Labs Leanstral 2603 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Voxtral Mini Transcribe Realtime 2602 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Voxtral Mini Realtime 2602 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Mistral Large 2411 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Mistral Large Pixtral 2411 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Large 2512 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Labs Mistral Small Creative | 60.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Mistral Small 2506 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Small 2603 | 60.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Mistral Large Latest | 40.0 | 5 | 8000.0% | 28821ms | 43647ms | 👁🔧 |
| mistral/mistral-large | 33.0 | 1 | 0.0% | 1284ms | 1284ms | |
| Glm 5 2 | 33.0 | 1 | 0.0% | 1200ms | 1200ms | 🔧 |
| Zai Glm 5 2 | 32.0 | 5 | 0.0% | 1149ms | 1632ms | 🔧 |
| Mistral: Mistral Small 3.1 24B (free) | 15.0 | 1 | 0.0% | 106306ms | 106306ms | 👁🔧 |
How to Use Mistral AI on Antbase
All Mistral AI models are available through Antbase's unified OpenAI-compatible API. You can either route directly to a specific Mistral AI model or let ANT's intelligent router automatically select the best model for your task.
python
from openai import OpenAI
client = OpenAI(
base_url="https://antbase.ai/v1",
api_key="your-ant-api-key",
)
# Let ANT auto-route (may use Mistral AI if optimal)
response = client.chat.completions.create(
model="ant:auto",
messages=[{"role": "user", "content": "Hello!"}],
)
# Or route directly to a Mistral AI model
response = client.chat.completions.create(
model="mistral/mistral-large",
messages=[{"role": "user", "content": "Hello!"}],
)Why Use Mistral AI Through Antbase?
- •No separate account needed — access Mistral AI models instantly through your Antbase API key.
- •Intelligent routing — ANT automatically selects the best Mistral AI model for each task, or falls over to another provider if Mistral AI is slow or down.
- •Same token prices — Antbase charges zero markup on Mistral AI token prices. You pay exactly what Mistral AI charges.
- •Automatic failover — if Mistral AI has an outage, your requests seamlessly route to an equivalent model on another provider.