Mistral Medium 3
nano-gpt/mistralai/mistral-medium-3 · family mistral-medium · in: text, image → out: text · price provider-declared (via models.dev). Capabilities: text image.
$0.80
Blended $/1M (3:1)
$0.4
Input $/1M
$2
Output $/1M
$2.40
1M in + 1M out
$0.2 −50%
Cache read $/1M
#3690 of 7,708
Cheapest rank
- Provider
- NanoGPT (
nano-gpt) - Kind
- text (outputs text)
- Context
- 131,072 tokens
- Max output
- 32,768 tokens
- Capabilities
- text image
Same model, other providers
“Mistral Medium 3” is offered by 9 providers — cheapest-effective first. Nvidia is the cheapest at $0.00 blended.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| Nvidia | free | free | $0.00 | 131k |
| NanoGPT (this) | $0.4 | $2 | $0.80 | 131k |
| OpenRouter | $0.4 | $2 | $0.80 | 131k |
| Azure | $0.4 | $2 | $0.80 | 128k |
| Azure Cognitive Services | $0.4 | $2 | $0.80 | 128k |
| Merge Gateway | $0.4 | $2 | $0.80 | 128k |
| Pioneer | $0.4 | $2 | $0.80 | 128k |
| Eden AI | $0.4 | $2 | $0.80 | 131k |
| Mistral | $0.4 | $2 | $0.80 | 131k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at NanoGPT and pass this model id:
curl $BASE_URL/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
-d '{"model":"mistralai/mistral-medium-3","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):