Ministral 8B
Mistral AI · released Oct 17, 2024
- Type
- Closed
- Input
- $0.11
- Output
- $0.11
- Cached
- None held
List price · per 1M tokens · Mistral AI at 128K context · source ↗
Our take
Written Sep 20, 2026Ministral 8B is a small text model you reach through a host rather than download, and its measured quality is thin: weak on graduate-level science and harder multi-task questions, stronger at following explicit instructions. Treat it as a cheap option for light chat and long-document work, not a base for demanding tasks.
Use it for light instruction-following and chat work where the bill matters more than peak quality, or for long documents that need not be split up first. Its instruction-following accuracy is the strongest thing measured about it. Skip it if your task needs measured science or multi-task reasoning, or if you meant to run the model on your own machine.
The case for it
- All three listed offers sit at the same rate for input and output, so the bill is predictable whichever host you pick.
- A 128,000-token request capacity means long documents need not be split up first, though reliable recall across all of it is unverified in our data.
- Instruction-following is its strongest measured area, at 59% on IFEval, which scores following explicit instructions rather than open-ended reasoning.
The case against it
- Weak on graduate-level science and harder multi-task questions: 4.6% correct on GPQA Diamond and 25.5% on MMLU-Pro.
- Its Arena figures are preference scores from human pairwise votes, not percentages, so they record which answer people preferred rather than whether it was correct.
- We list no download for it, so using it means choosing a host, and no licence is supplied, so permissions are unverified in our data.
How good is it?
A small closed text model for chat and general requests, though it trails most others at everyday questions, writing and code.
- getting answers to everyday questionsArena Text (overall) · 163rd of 168
- drafts, rewrites and editingArena Creative Writing · 155th of 168
- writing and completing codeArena Coding · 164th of 168
EverydayGeneral questions and everyday reasoning
Arena Text (overall)163rd of 168 · 1238
CodingWriting and fixing code on its own
Arena Coding164th of 168 · 1274
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing155th of 168 · 1222
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Where to rent it
Prices checked between 35 days and 38 days ago — each listing carries its own date.
Mistral AI, through OpenRouter
The lab is also the cheapest we hold. The strip above and this offer are the same one, so nothing on this page undercuts Mistral AI on 128K of context.
- per 1M tokens
- $0.11 in / $0.11 out
- Context served
- 128K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $0.11 / $0.11checked 35 days ago | 128K | not measured | Unknown | Unknown | Unknown |
| Mistral AIThrough OpenRouter | $0.11 / $0.11checked 38 days ago | 128K | not measured | No | Yes30 days | Unknown |
| Mistral AIusThrough OpenRouter | $0.11 / $0.11checked 35 days ago | 128K | not measured | No | Yes30 days | Unknown |
OpenRouterAcross the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 does not say. 0 appear in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✗ | ✓ | ✓ |
| Mistral AIThrough OpenRouter | ✗ | ✓ | ✓ |
| Mistral AIusThrough OpenRouter | ✗ | ✓ | ✓ |
Tool calling: 0 of 3 listings say yes, 3 say no. JSON output: 3 of 3 listings say yes. Strict schema: 3 of 3 listings say yes.
Models people weigh against Ministral 8B
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 listings does not say whether it trains on prompts.
- We hold no cached-input rate for any of its listings.
- We hold no batch or off-peak rate for any of its listings.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
We hold no licence record for this model, and no record of published weights either — so we can neither summarise its terms nor point you at the weights.
Identifiers
- Takes in, gives back
- Text in, text out
- Catalogue slug
- mistralai-ministral-8b