Model comparison
Ministral 3 (3B / 8B / 14B) vs Ministral 3B / 8B (24.10)
- Input
- $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens
- Output
- $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens
- Context
- 256K
- Released
- 2025-12-02
- Input
- $0.04 / M tokens (3B); $0.10 / M tokens (8B)
- Output
- $0.04 / M tokens (3B); $0.10 / M tokens (8B)
- Context
- 128K tokens
- Released
- 2024-10-09
Ministral 3B / 8B (24.10) is cheaper on input tokens ($0 vs $0.1 per 1M, 2.5×); Ministral 3 (3B / 8B / 14B) has the larger context window (256K vs 128K).
Specs and pricing side by side
| Spec | Ministral 3 (3B / 8B / 14B) | Ministral 3B / 8B (24.10) |
|---|---|---|
| Maker | Mistral AI | Mistral AIsame |
| Released | 2025-12-02 | 2024-10-09 |
| Context window | 256K | 128K tokens |
| Max output | not published | Not separately published by Mistral; output… |
| Input price | $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens | $0.04 / M tokens (3B); $0.10 / M tokens (8B) |
| Output price | $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens | $0.04 / M tokens (3B); $0.10 / M tokens (8B) |
| Parameters | 3B, 8B, 14B (dense) | Ministral 3B: ~3B. Ministral 8B: ~8B (8,019,808,256 parameters). |
| License | Apache 2.0 | Ministral 8B: Mistral Research License (open weights for research; commercial use requires a separate Mistral commercial license). Ministral 3B: Mistral Commercial License only — weights were not released openly. |
| Weights | Open weights | Open weightssame |
| Modalities | Text, Vision | text input, text output |
| Knowledge cutoff | Not disclosed by Mistral AI | Not officially disclosed by Mistral AI. |
Benchmarks both models report
These two models have no benchmark in common in our registry, so there is no like-for-like score to compare. See each model’s own results: Ministral 3 (3B / 8B / 14B), Ministral 3B / 8B (24.10).
Which one should you pick?
Pick Ministral 3 (3B / 8B / 14B) if…
- You need the longer context (256K vs 128K).
- You need Text / Vision input, which Ministral 3B / 8B (24.10) lacks.
- Your use case is: on-device and offline AI assistants on laptops, phones, and embedded hardware.
Pick Ministral 3B / 8B (24.10) if…
- Cost matters: input tokens cost 2.5× less ($0 vs $0.1 per 1M).
- You need text input / text output input, which Ministral 3 (3B / 8B / 14B) lacks.
- Your use case is: on-device and edge inference on phones, laptops and IoT devices.
FAQ
Which is better, Ministral 3 (3B / 8B / 14B) or Ministral 3B / 8B (24.10)?
Ministral 3B / 8B (24.10) is cheaper on input tokens ($0 vs $0.1 per 1M, 2.5×); Ministral 3 (3B / 8B / 14B) has the larger context window (256K vs 128K). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is Ministral 3B / 8B (24.10)?
Ministral 3 (3B / 8B / 14B): $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens input, $0.10 (3B) / $0.15 (8B) / $0.20 (14B) per 1M tokens output. Ministral 3B / 8B (24.10): $0.04 / M tokens (3B); $0.10 / M tokens (8B) input, $0.04 / M tokens (3B); $0.10 / M tokens (8B) output (list API prices as published by each maker).
What are the context windows of Ministral 3 (3B / 8B / 14B) and Ministral 3B / 8B (24.10)?
Ministral 3 (3B / 8B / 14B) offers 256K; Ministral 3B / 8B (24.10) offers 128K tokens with up to Not separately published by Mistral; output shares the 128K-token window. output.
Is Ministral 3 (3B / 8B / 14B) newer than Ministral 3B / 8B (24.10)?
Ministral 3 (3B / 8B / 14B) was released 2025-12-02; Ministral 3B / 8B (24.10) was released 2024-10-09.