Ministral 3 8B 2512 (batch)
Mistral · Multimodal · Released Dec 2025
An 8-billion parameter language model from Mistral with vision capabilities and a 262K token context window, positioned between the smaller Ministral 3 3B and larger Ministral 3 14B in the family.
- Strengths
- Balances model capacity with inference speed, handling both text and image inputs efficiently within its parameter budget.
- Best for
- Applications requiring moderate reasoning and vision understanding where latency and throughput matter more than frontier capability, or as a stepping stone from smaller models like Ministral 3 3B.
- Limitations
- Smaller than Ministral 3 14B and the larger Mistral models, so it trades raw reasoning capacity for efficiency and may not match frontier performance on complex tasks.
Input / 1M
$0.075
Output / 1M
$0.075
Cached input / 1M
$0.0075
Context window
262K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in decreased 50.0% out decreased 50.0%
- 90d
- in decreased 50.0% out decreased 50.0%
- 1y
- in decreased 50.0% out decreased 50.0%
- Since launch
- in decreased 50.0% out decreased 50.0%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 10 Sep 2026 | $0.075 | $0.075 | $0.0075 | Imported from OpenRouter | openrouter.ai |
| 28 Aug 2026 | $0.15 | $0.15 | $0.015 | Imported from OpenRouter | openrouter.ai |