$ diff mistral-large-4 granite-4-2
Mistral Large 4 vs Granite 4.2
Values come from Mistral Large 4 and Granite 4.2, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing.
Mistral AIMistral Large 4
- context
- 1M
- weights
- —
- $/M in
- $1.36
- $/M out
- $4.18
- context
- 512K
- weights
- open
- $/M in
- —
- $/M out
- —
What actually differs
- Context window
- Mistral Large 4 takes 1M against 512K — modestly more room in a single request.
- Recency
- Mistral Large 4 shipped 42 days after Granite 4.2 (2026-10-06 vs 2026-08-25).
Full spec
| Attribute | Mistral Large 4 | Granite 4.2 |
|---|---|---|
| Developer | Mistral AI | IBM |
| Released | 2026-10-06 | 2026-08-25 |
| Context window | 1M | 512K (30B); unknown (3B, 8B) |
| Pricing | $1.36/M input · $4.18/M output (announcement) | not applicable — open weights, no hosted rate published |
| License | not recorded | Apache 2.0 |
| Availability | Mistral Studio API preview; weights promised by end of October 2026 | Hugging Face, Ollama, GitHub |