Ministral 3B (latest)
MistralOpen weights
Ministral 3B (latest) by Mistral costs $0.10 per 1M input tokens and $0.10 per 1M output tokens, with a 131K-token context window.
Pricing
Input (per 1M tokens)
$0.10
Output (per 1M tokens)
$0.10
Cached input (per 1M)
$0.01
Specifications
- Provider
- Mistral
- Context window
- 131K tokens
- Parameters
- —
- Released
- Oct 2024
- Open weights
- Yes
- Frontier model
- No
Benchmarks
- GPQA Diamond
- 25.3
Speed & latency
Time to first token
0.33s
Output speed
207 tok/s
Fastest host: mistral (via vercel-ai-gateway).
Compare Ministral 3B (latest) with…
Ministral 3B (latest) vs Gemma 3 4B$0.10/1MMinistral 3B (latest) vs Qwen2.5-Coder-0.5B$0.10/1MMinistral 3B (latest) vs Granite 4.1 8B$0.10/1MMinistral 3B (latest) vs DeepSeek R1 0528 Qwen3 8B$0.09/1MMinistral 3B (latest) vs Granite 4.0 Micro$0.11/1MMinistral 3B (latest) vs gpt-oss-20b (batch)$0.11/1M
FAQ
How much does Ministral 3B (latest) cost?
Ministral 3B (latest) is priced at $0.10 per 1M input tokens and $0.10 per 1M output tokens. Output tokens cost more than input.
What is Ministral 3B (latest)'s context window?
Ministral 3B (latest) supports a context window of 131K tokens — the maximum prompt plus response length it can handle in a single request.
Is Ministral 3B (latest) open weight?
Yes — Ministral 3B (latest) has openly available weights.
Data as of 2026-11-01. Source: models.dev (open). Pricing is per 1M tokens (USD); confirm with the provider before production use.