Mistral AI model product
Mistral Medium 3.5
5 is a dense 128B-parameter multimodal model with a 256k context window for instruction-following, reasoning, and coding.
Updated Sep 8, 2026. Default version: Mistral Medium 3.5
LLMBoard Score47.6Mistral Medium 3.5
Coverage0%8 benchmark families
Context window256KTokens
Official input price$1.5Mistral API
Mistral Medium 3.5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Mistral Medium 3.5 Benchmark Results
Benchmark scores for Mistral Medium 3.5.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkTau3 Airline | Score72.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTau3 Retail | Score76.10% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTau3 Telecom | Score91.40% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkBeyond AIME | Score66.90% | Rank04 | Participants5 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkCOLLIE | Score95.80% | Rank04 | Participants10 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkTau3 Banking | Score13.40% | Rank05 | Participants7 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkFinance Agent v2 | Score32.10% | Rank22 | Participants26 | Percentile16.00% | EvidenceB | Evaluated |
| BenchmarkSWE-Bench Verified | Score77.60% | Rank26 | Participants113 | Percentile77.68% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-0.50% | Rank27 | Participants49 | Percentile45.83% | EvidenceA | Evaluated |
| BenchmarkIFBench | Score69.00% | Rank29 | Participants39 | Percentile26.32% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score-6.13% | Rank37 | Participants49 | Percentile25.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score-7.16% | Rank42 | Participants49 | Percentile14.58% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score-12.61% | Rank43 | Participants49 | Percentile12.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score-14.49% | Rank45 | Participants49 | Percentile8.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score-2.08% | Rank48 | Participants49 | Percentile2.08% | EvidenceA | Evaluated |
| BenchmarkBrowseComp | Score48.60% | Rank50 | Participants63 | Percentile20.97% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,222.18 rating | Rank56 | Participants107 | Percentile48.11% | EvidenceA | Evaluated |
| BenchmarkAIME 2025 | Score86.30% | Rank62 | Participants119 | Percentile48.31% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,199.04 rating | Rank65 | Participants107 | Percentile39.62% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,424.74 rating | Rank75 | Participants210 | Percentile64.59% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,430.42 rating | Rank79 | Participants121 | Percentile35.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,426.87 rating | Rank79 | Participants210 | Percentile62.68% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,265.03 rating | Rank86 | Participants97 | Percentile11.46% | EvidenceA | Evaluated |
Mistral Medium 3.5 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categorycreative writing | Rank14 | Rating / score1,223.06 | Votes1,365 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing | Rank15 | Rating / score1,204.12 | Votes1,365 | ObservationsN/A | Result date |
| Arenaagent steerability | Categoryoverall | Rank27 | Rating / score0.00 | VotesN/A | Observations5.5K | Result date |
| Arenavision | Categorycaptioning | Rank27 | Rating / score1,144.95 | Votes403 | ObservationsN/A | Result date |
| Arenavision style control | Categorycaptioning | Rank27 | Rating / score1,129.45 | Votes403 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank29 | Rating / score1,402.00 | Votes588 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank36 | Rating / score1,449.83 | Votes170 | ObservationsN/A | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank37 | Rating / score-0.06 | VotesN/A | Observations11.2K | Result date |
| Arenatext factuality | Categoryfrench | Rank37 | Rating / score1,463.30 | Votes1,055 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank37 | Rating / score1,367.66 | Votes485 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank37 | Rating / score1,401.93 | Votes1,076 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryspanish | Rank40 | Rating / score1,432.10 | Votes1,212 | ObservationsN/A | Result date |
| Arenavision | Categoryentity recognition | Rank40 | Rating / score1,148.48 | Votes433 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank40 | Rating / score1,129.65 | Votes433 | ObservationsN/A | Result date |
| Arenaagent | Categoryoverall | Rank42 | Rating / score-0.07 | VotesN/A | Observations294.2K | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank42 | Rating / score1,473.14 | Votes2,588 | ObservationsN/A | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank43 | Rating / score-0.13 | VotesN/A | Observations1.8K | Result date |
| Arenatext | Categoryfrench | Rank43 | Rating / score1,460.53 | Votes2,037 | ObservationsN/A | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank45 | Rating / score-0.14 | VotesN/A | Observations5.1K | Result date |
| Arenavision | Categorychinese | Rank46 | Rating / score1,268.38 | Votes244 | ObservationsN/A | Result date |
| Arenavision | Categorydiagram | Rank46 | Rating / score1,253.87 | Votes1,471 | ObservationsN/A | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank48 | Rating / score-0.02 | VotesN/A | Observations270.5K | Result date |
| Arenavision | Categoryenglish | Rank50 | Rating / score1,241.00 | Votes2,224 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank50 | Rating / score1,245.42 | Votes244 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank51 | Rating / score1,209.36 | Votes224 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycoding | Rank52 | Rating / score1,493.59 | Votes3,059 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank53 | Rating / score1,460.75 | Votes3,077 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank53 | Rating / score1,442.72 | Votes170 | ObservationsN/A | Result date |
| Arenavision style control | Categorydiagram | Rank53 | Rating / score1,241.18 | Votes1,471 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank53 | Rating / score1,189.70 | Votes224 | ObservationsN/A | Result date |
Mistral Medium 3.5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.5 input, $7.5 output per 1M
- Official provider
- Mistral
- Lowest third-party
- From $1.5 input, $7.5 output per 1M via Eden AI
- Tracked offerings
- 3
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderMistral | Provider model IDmistral-medium-2604 | Regionglobal | Input / 1M$1.5 | Output / 1M$7.5 | Context262.1K | Updated |
| ProviderEden AI | Provider model IDmistral/mistral-medium-2604 | Regionglobal | Input / 1M$1.5 | Output / 1M$7.5 | Context262.1K | Updated |
| ProviderRequesty | Provider model IDmistral-medium-3-5 | Regionglobal | Input / 1M$1.65 | Output / 1M$8.25 | Context262.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Mistral Medium 3.5 Runtime Performance
Provider-specific output speed and catalog latency for Mistral Medium 3.5. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Mistral Medium 3.5 Specifications
Technical details for the model's default version.
- Version
- Mistral Medium 3.5
- Released
- Apr 29, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 128B
- Context window
- 256K
- Max output
- 256K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Modified MIT License
Mistral Medium 3.5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionMistral Medium 3.5 | Released | LLMBoard47.56 | Parameters128B | Context256K | Max output256K | Open weightsYes | LicenseModified MIT License |
Models similar to Mistral Medium 3.5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Mistral Small 4
Mistral AI
MA
Ministral 3
Mistral AI
MA
Devstral Medium
Mistral AI
MA
Pixtral Large
Mistral AI
MA
Min istral 3
Mistral AI
MA
Magistral Small
Mistral AI
What is Mistral Medium 3.5?
Key information about Mistral Medium 3.5 and its available data.
5 is Mistral AI's merged model, released as open weights under a Modified MIT License. It supports configurable reasoning effort and uses a vision encoder trained from scratch for variable image sizes and aspect ratios.
Data as of 2026-09-08.
FAQ
Common questions about Mistral Medium 3.5.
When was Mistral Medium 3.5 released?
Mistral Medium 3.5's default version was released on Apr 29, 2026.
How much does Mistral Medium 3.5 cost?
Mistral Medium 3.5's official API price is $1.5 per million input tokens and $7.5 per million output tokens via Mistral. The lowest tracked third-party offer starts at $1.5 input and $7.5 output via Eden AI.
Who created Mistral Medium 3.5?
Mistral Medium 3.5 was created by Mistral AI.
What is the context window for Mistral Medium 3.5?
The default version has a 256K token context window.
Is Mistral Medium 3.5 open weight?
Yes. The default version is marked as open weight under Modified MIT License.