Mistral AI model product
Ministral 8B is an instruct fine-tuned language model for local, on-device, and edge computing use cases.
Updated Sep 8, 2026. Default version: Ministral 8B Instruct
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Ministral 8B Instruct.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkFrench MMLU | Score57.50% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMBPP pass@1 | Score70.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAGIEval | Score48.30% | Rank09 | Participants10 | Percentile11.11% | EvidenceC | Evaluated |
| BenchmarkMT-Bench | Score83.00% | Rank09 | Participants12 | Percentile27.27% | EvidenceC | Evaluated |
| BenchmarkArena Hard | Score70.90% | Rank11 | Participants26 | Percentile60.00% | EvidenceC | Evaluated |
| BenchmarkWinogrande | Score75.30% | Rank13 | Participants22 | Percentile42.86% | EvidenceC | Evaluated |
| BenchmarkTriviaQA | Score65.50% | Rank15 | Participants18 | Percentile17.65% | EvidenceC | Evaluated |
| BenchmarkARC-C | Score71.90% | Rank18 | Participants34 | Percentile48.48% | EvidenceC | Evaluated |
| BenchmarkMATH | Score54.50% | Rank52 | Participants71 | Percentile27.14% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score34.80% | Rank66 | Participants66 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score65.00% | Rank92 | Participants101 | Percentile9.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,190.70 rating | Rank201 | Participants210 | Percentile4.31% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,237.28 rating | Rank203 | Participants210 | Percentile3.35% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categorychinese | Rank191 | Rating / score1,193.27 | Votes478 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank191 | Rating / score1,188.21 | Votes683 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank193 | Rating / score1,206.26 | Votes703 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank193 | Rating / score1,184.02 | Votes543 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank193 | Rating / score1,256.00 | Votes478 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank193 | Rating / score1,213.50 | Votes683 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank195 | Rating / score1,211.78 | Votes720 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank195 | Rating / score1,221.07 | Votes678 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank196 | Rating / score1,202.21 | Votes838 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank196 | Rating / score1,169.87 | Votes332 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank196 | Rating / score1,213.59 | Votes1,252 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank197 | Rating / score1,191.11 | Votes1,257 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank197 | Rating / score1,216.90 | Votes543 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank198 | Rating / score1,174.56 | Votes678 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank198 | Rating / score1,237.29 | Votes332 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry software and it services | Rank198 | Rating / score1,276.74 | Votes1,252 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank199 | Rating / score1,211.11 | Votes324 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank199 | Rating / score1,265.15 | Votes703 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry business and management and financial operations | Rank199 | Rating / score1,234.95 | Votes572 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank199 | Rating / score1,269.86 | Votes324 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank199 | Rating / score1,245.13 | Votes760 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank200 | Rating / score1,172.31 | Votes572 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank200 | Rating / score1,179.46 | Votes280 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank200 | Rating / score1,194.43 | Votes760 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank200 | Rating / score1,274.45 | Votes838 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank201 | Rating / score1,060.18 | Votes3,020 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank201 | Rating / score1,190.70 | Votes4,781 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts | Rank201 | Rating / score1,250.50 | Votes1,257 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank201 | Rating / score1,225.31 | Votes1,350 | ObservationsN/A | Result date |
| Arenatext style control | Categorylonger query | Rank201 | Rating / score1,258.91 | Votes720 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderPioneer | Provider model IDmistralai/Ministral-8B-Instruct-2410 | Regionglobal | Input / 1M$0.15 | Output / 1M$0.15 | Context128K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Ministral 8B Instruct. Runtime does not affect the capability score.
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderMistral AI | Output Speed0.10 tok/s | Catalog Latency0.18 s | Max Input128K | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionMinistral 8B Instruct | Released | LLMBoard0.00 | Parameters8B | Context128K | Max output128K | Open weightsYes | LicenseMistral Research License |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Ministral 8B and its available data.
Ministral 8B is an instruct fine-tuned model from Mistral AI, designed for local intelligence, on-device computing, and at-the-edge use cases.
Data as of 2026-09-08.
Common questions about Ministral 8B.
Ministral 8B's default version was released on Oct 16, 2024.
No official standard PAYG price is currently available for Ministral 8B. The lowest tracked third-party offer starts at $0.15 input and $0.15 output via Pioneer.
Ministral 8B was created by Mistral AI.
The default version has a 128K token context window.
Yes. The default version is marked as open weight under Mistral Research License.
1 provider offerings are linked to the default version.