// Mistral AI
Ministral 3 8B
A Mistral AI model. 1 prices across 1 channels side by side, each with its source and check date – plus specs, licence, speed and benchmarks.
- Official in/out per 1M
- $0.15 / $0.15
- Lowest (official)
- $0.15 / $0.15
- Context
- 262K
- AA index
- 5.5
Prices checked 2026-10-10Official sources linked on every rowNo estimates
01Specs at a glance
Official values first; gaps filled from OpenRouter’s API where the vendor page is silent.
| Vendor | Mistral AI |
|---|---|
| Context window | 262K (262,144 tokens) |
| Max output | 210K |
| Input | Text · Image |
| Output | Text |
| Tool calling / JSON | Yes / Yes |
| Weights & licence | Open weights · Apache-2.0 · Hugging Face |
| Parameters | 8.9B (from weight files) |
| Released | Dec 2, 2025 (OpenRouter listing date) |
| Model IDs | ministral-3-8b |
02Prices on 1 channels
Per 1M tokens, in the currency each source publishes (1 USD = 6.7 CNY for the blended column). Sorted: official first, then cheapest.
| Channel | Source | Input | Output | Cache hit | Blended $ | Context | Notes |
|---|---|---|---|---|---|---|---|
| Official Vendor API, billed in USD | Official 10-10 | $0.15 | $0.15 | $0.015 | 0.15 | 262K | Batch −50%Batch -50% (pricing FAQ) |
“Official” = vendor pricing page; “Platform” = the platform’s own pricing page; “OpenRouter” = price listed on OpenRouter, not checked on the platform’s site. Click a source tag to open the original page.
03Where you can use it
Derived only from the price sources above. Country restrictions are set by each vendor – check their terms before you sign up.
- Vendor API, billed in USD
Official - Self-host the open weights
GPU / VRAM sizing and build plans→
04Performance
All results →Speed and AA index from Artificial Analysis public leaderboards (not our own tests); benchmark scores from our benchmark library, best result per benchmark.
- AA indexMinistral 3 8B
- 5.5
- Output speed (median)
- 111 tok/s
- Time to first token
- 0.8s
- AA index per $ (blended)
- 36.52
| Benchmark | Best score | Source | Date |
|---|---|---|---|
| LMArena Text | 1238 | official | 2026-10-08 |
| Open LLM Leaderboard v2 | 24.2% | official | 2025-03-13 |
| GPQA Diamond | 27.1% | independent | 2025-01-27 |
| MMLU-Pro | 37.9% | official | 2026-03-11 |
| MATH Level 5 | 14.9% | independent | 2025-01-27 |
| BFCL v4 (function calling) | 11.1% | official | 2026-10-17 |
05Price changes we recorded
All price changes →From our daily re-checks of the sources above.
No price change recorded since we started tracking (2026-10-10).
06More from Mistral AI
FAQ
How much does the Ministral 3 8B API cost?
On the official pricing page: $0.15 / $0.15 per 1M input / output tokens, cached input $0.015 (checked 2026-10-10).
What is the context window of Ministral 3 8B?
262K tokens of context, up to 210K output tokens. Platforms may cap it lower – see the table.
Is Ministral 3 8B open source?
Yes, the weights are published on Hugging Face (mistralai/Ministral-3-8B-Instruct-2512). Licence: Apache-2.0.