VerdictPal
Join

Models / Mistral

Mistral Medium 3.5

Mistral Medium · Released 2026-04-29

AA Intelligence Index 29.9. Listed on Artificial Analysis with API pricing $1.5/$7.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#153 of 184 on AA Index · current snapshot
15 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite43
ContextUnlisted
Input / 1M$1.5
Output / 1M$7.5
Knowledge cutoffunpublished
Output speed158 t/s
TTFT (default API)0.67s
Statussolid

Family profile

Best published score in each covered benchmark family.

59/100 avg
Reasoning
75
Coding
40
Agentic
51
Long context
69

4 tested benchmark families

Benchmark placements

Where Mistral Medium 3.5 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
AA output speedPerformance#15/ 190Artificial Analysis2026-09-10
IFBenchReasoning#58/ 12769Artificial Analysis2026-09-10
AA time to first tokenPerformance#74/ 190Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#77/ 11015Artificial Analysis2026-09-10
SciCodeCoding#78/ 9340Artificial Analysis2026-09-10
Terminal-BenchAgentic#95/ 18951Artificial Analysis2026-09-10
AA-LCRLong context#126/ 18269Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#146/ 18514Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#158/ 18815Artificial Analysis2026-09-10
GPQA DiamondReasoning#159/ 19075Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Mistral Medium 3.5 shows up, it's highlighted.

ReasoningMistral Medium 3.5 · 75
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
CodingMistral Medium 3.5 · 40
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticMistral Medium 3.5 · 51
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextMistral Medium 3.5 · 69
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Mistral Medium 3.5 in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context69.33%Artificial Analysis6
AA output speedPerformance158 t/sArtificial Analysis6
AA time to first tokenPerformance0.67sArtificial Analysis6
Artificial Analysis Intelligence IndexReasoning14.9%Artificial Analysis6
GPQA DiamondReasoning74.8%Artificial Analysis6
Humanity's Last ExamReasoning13.8%Artificial Analysis6
IFBenchReasoning68.78%Artificial Analysis6
SciCodeCoding40.2%Artificial Analysis6
τ³-Bench BankingAgentic15.05%Artificial Analysis6
Terminal-BenchAgentic50.56%Artificial Analysis6
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning75
Coding40
Agentic51
Long context69
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Mistral Medium 3.5. Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 37 catalog benchmarks have a sourced row for Mistral Medium 3.5.

Reasoning

Coding

Agentic

Long context

Performance

Sources

The Pack · Editorial newsletter

New tools in your inbox. Free.

One short email when a tool ships or changes status. No tracking, no third-party analytics. Unsubscribe in one click.