Mistral AI API Models & Token Pricing
Tracking 21 Mistral AI endpoints from $0.029 blended/1M up to 1.05M context.
Models Tracked
21
Cheapest Paid
$0.029
Average Blended
$1.03
Max Context
1.05M
| Model | Input / 1M | Output / 1M | Blended | Context |
|---|---|---|---|---|
|
Mistral: Mistral Nemo
Open Weight
|
$0.029 | $0.03 | $0.029 | 131K |
|
Mistral: Mistral Small 3
ultra-budget
|
$0.05 | $0.08 | $0.058 | 33K |
|
Mistral: Ministral 3 3B 2512
Vision
|
$0.10 | $0.10 | $0.10 | 131K |
|
Mistral: Mistral Small 3.2 24B
Vision
|
$0.075 | $0.20 | $0.106 | 256K |
|
Mistral: Ministral 3 8B 2512
Vision
|
$0.15 | $0.15 | $0.15 | 262K |
|
Mistral: Voxtral Small 24B 2507
ultra-budget
|
$0.10 | $0.30 | $0.15 | 33K |
|
Mistral: Ministral 3 14B 2512
Vision
|
$0.20 | $0.20 | $0.20 | 262K |
|
Mistral: Mistral Small 4
Vision
|
$0.15 | $0.60 | $0.263 | 262K |
|
Mistral: Saba
ultra-budget
|
$0.20 | $0.60 | $0.30 | 33K |
|
Mistral: Mistral Small 3.1 24B
Vision
|
$0.351 | $0.555 | $0.402 | 128K |
|
Mistral: Codestral 2508
budget
|
$0.30 | $0.90 | $0.45 | 256K |
|
Mistral: Mistral Large 3 2512
Vision
|
$0.50 | $1.50 | $0.75 | 262K |
|
Mistral: Devstral 2 2512
budget
|
$0.40 | $2.00 | $0.80 | 262K |
|
Mistral: Mistral Medium 3
Vision
|
$0.44 | $2.20 | $0.88 | 131K |
|
Mistral: Mistral Medium 3.1
Vision
|
$0.44 | $2.20 | $0.88 | 131K |
|
Mistral: Mistral Large 4
Vision
|
$0.70 | $2.10 | $1.05 | 1.05M |
|
Mistral: Mistral Medium 3.5
Vision
|
$1.50 | $7.50 | $3.00 | 262K |
|
Mistral Large 2407
balanced
|
$2.00 | $6.00 | $3.00 | 131K |
|
Mistral Large
balanced
|
$2.00 | $6.00 | $3.00 | 128K |
|
Mistral: Pixtral Large 2502
Vision
|
$2.00 | $6.00 | $3.00 | 128K |
|
Mistral: Mixtral 8x22B Instruct
Open Weight
|
$2.00 | $6.00 | $3.00 | 66K |
Access Mistral AI Endpoints via Partner Gateways
Multi-Provider Gateway
Call via OpenRouter →
OpenRouter
Unified API gateway for 300+ LLMs with automatic fallbacks and unified billing.
Open-Weight Inference
Run on Together AI →
Together AI
High-throughput serverless inference & fine-tuning for Llama, DeepSeek, and Qwen.
Low-Latency Engine
Deploy on Fireworks →
Fireworks AI
Ultra-low-latency inference engine built for production structured outputs and function calling.
Some outbound infrastructure and API gateway links on KostoZero are partner referrals that help keep this observatory free and independent at no extra cost to you.
6h Autopilot Radar
Get Instant Price-Drop (>10%), New AI Bot & SaaS Cliff Alerts
Every 6 hours our autonomous sync checks 40+ LLMs, AI crawlers, and SaaS pricing cliffs—alerting you when prices drop >10% or new scrapers appear.