Enter opens the highlighted result. Esc closes.

Mistral Medium 3.1 vs Hy3

Hy3 costs 6.1x less per output token than Mistral Medium 3.1 ($0.33 vs $2.00 per million). Updated October 21, 2026; 41 benchmarks have a result for both.

Replace Mistral Medium 3.1 Replace Hy3
Mistral Medium 3.1Mistral AI
Coding
No data
Overall
No data
Output $/M
$2.00
2
Hy3Tencent
Coding
No data
Overall
No data
Output $/M
$0.33
39
vs boards won of 41 shared

Mistral Medium 3.1 or Hy3: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick Mistral Medium 3.1 for work measured by SciCode or ALE-Bench, where it leads.

Model AMistral Medium 3.1Mistral AINot benchmarkedNo coding benchmarkNo cost per solved taskModel BHy3TencentNot benchmarked1 independent benchmarkNo cost per solved task
Shared results
41
Mistral Medium 3.1 only
33
Hy3 only
3
Like-for-like categories
6 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Mistral Medium 3.1 not benchmarked, Hy3 not benchmarked.
Widest category
Web design, 15.3 points apart on average over 8 shared benchmarks.
Widest single gap
Long context reasoning (AA-LCR): Mistral Medium 3.1 42.7% against Hy3 79.0%, 36.3 points apart on a 0 to 100 scale.
Output price
Mistral Medium 3.1 $2.00 against Hy3 $0.33 per million output tokens, 6.1x apart.
Context window
Mistral Medium 3.1 takes 131K tokens, Hy3 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryMistral Medium 3.1BarsHy3Ties
Coding04
Reasoning05
Math02
Knowledge13
Web design08
General117
All shared boards239
CodingLike-for-likeSix of these boards feed the Vibe Coding Index04
Coding benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
LMArena WebDevNot measured1507Only Hy3 measured
AA Coding Index20.5Not measuredOnly Mistral Medium 3.1 measured
LiveCodeBench (Vals AI)44.8%Not measuredOnly Mistral Medium 3.1 measured
LiveCodeBench (Artificial Analysis run)40.6%Not measuredOnly Mistral Medium 3.1 measured
SciCode32.4%48.6%Hy3
Terminal-Bench Hard (Artificial Analysis run)10.6%Not measuredOnly Mistral Medium 3.1 measured
ALE-Bench210.2838.6Hy3
LMArena Text: coding14331466Hy3
LMArena Text: software and IT14361463Hy3
LMArena WebDev: ReactNot measured1511Only Hy3 measured
LMArena WebDev: HTMLNot measured1487Only Hy3 measured
Neither model has a result on the other 28 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Intelligence benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
AA Intelligence Index9.2Not measuredOnly Mistral Medium 3.1 measured
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Agentic work benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
AA Agentic Index1.9Not measuredOnly Mistral Medium 3.1 measured
τ²-Bench Telecom40.6%Not measuredOnly Mistral Medium 3.1 measured
Neither model has a result on the other 39 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
MMMU Pro (Vals AI)63.0%Not measuredOnly Mistral Medium 3.1 measured
GPQA Diamond (Artificial Analysis run)58.8%89.7%Hy3
Humanity's Last Exam (Artificial Analysis run)4.7%33.5%Hy3
LMArena Text: hard prompts14261447Hy3
LMArena Text: expert14081461Hy3
LMArena Text: hard prompts in English14351455Hy3
LMArena Vision: diagrams1175Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: homework1165Not measuredOnly Mistral Medium 3.1 measured
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
AIME42.3%Not measuredOnly Mistral Medium 3.1 measured
MATH 500 (Vals AI)87.0%Not measuredOnly Mistral Medium 3.1 measured
MGSM (Vals AI)90.9%Not measuredOnly Mistral Medium 3.1 measured
AA Math Index38.3Not measuredOnly Mistral Medium 3.1 measured
AIME 2025 (Artificial Analysis run)38.3%Not measuredOnly Mistral Medium 3.1 measured
LMArena Text: math14081473Hy3
LMArena Text: mathematical work14121482Hy3
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
Knowledge benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
MMLU Pro75.3%Not measuredOnly Mistral Medium 3.1 measured
LegalBench (Vals AI)62.0%Not measuredOnly Mistral Medium 3.1 measured
CorpFin (Vals AI)50.7%Not measuredOnly Mistral Medium 3.1 measured
TaxEval (Vals AI)70.3%Not measuredOnly Mistral Medium 3.1 measured
MortgageTax (Vals AI)36.5%Not measuredOnly Mistral Medium 3.1 measured
MedQA (Vals AI)78.2%Not measuredOnly Mistral Medium 3.1 measured
MMLU Pro (Artificial Analysis run)68.3%Not measuredOnly Mistral Medium 3.1 measured
LMArena Text: science14441457Hy3
LMArena Text: medicine and healthcare14491452Hy3
LMArena Text: legal and government14391422Mistral Medium 3.1
LMArena Text: business and finance14181424Hy3
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index08
Web design benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
Design Arena: websites11331230Hy3
Design Arena: UI components11001207Hy3
Design Arena: game dev10741208Hy3
Design Arena: data viz11441163Hy3
Design Arena: 3D10811233Hy3
Design Arena: all code categories11201227Hy3
Design Arena: SVG9931200Hy3
Design Arena: ASCII art9971152Hy3
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index117
General benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
IFBench39.8%Not measuredOnly Mistral Medium 3.1 measured
Long context reasoning (AA-LCR)42.7%79.0%Hy3
DTBench67.5%Not measuredOnly Mistral Medium 3.1 measured
LMCA19.1%Not measuredOnly Mistral Medium 3.1 measured
LMArena Text14241438Hy3
LMArena Text: instruction following13981426Hy3
LMArena Text: creative writing13911403Hy3
LMArena Vision1172Not measuredOnly Mistral Medium 3.1 measured
LMArena Text: English14381448Hy3
LMArena Text: non-English14081426Hy3
LMArena Text: Chinese14471485Hy3
LMArena Text: German14321449Hy3
LMArena Text: Japanese13791397Hy3
LMArena Text: Korean13801396Hy3
LMArena Text: French14601456Mistral Medium 3.1
LMArena Text: Spanish14331457Hy3
LMArena Text: Russian14111432Hy3
LMArena Text: Polish14301447Hy3
LMArena Text: multi-turn14181436Hy3
LMArena Text: longer queries14061442Hy3
LMArena Text: writing and language13951414Hy3
LMArena Text: entertainment and media13901400Hy3
LMArena Vision: English1179Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: Chinese1190Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: reading text1175Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: humor1168Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: captioning1145Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: recognizing things1147Not measuredOnly Mistral Medium 3.1 measured
LMArena Vision: creative writing1200Not measuredOnly Mistral Medium 3.1 measured
Neither model has a result on the other 17 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Mistral Medium 3.1 and Hy3
BenchmarkMistral Medium 3.1Hy3Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Mistral Medium 3.1$0.0014Fits in one request
Hy3$0.0002Fits in one request

Hy3 costs less

Repository review

50K input + 3K output tokens

Mistral Medium 3.1$0.026Fits in one request
Hy3$0.0051Fits in one request

Hy3 costs less

Agent loop

200K input + 10K output tokens

Mistral Medium 3.1$0.100Exceeds the context window
Hy3$0.020Fits in one request

Hy3 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Mistral Medium 3.1 and Hy3
SpecificationMistral Medium 3.1Tencent Hy3
MakerMistral AITencent
WeightsNo public weights listedOpen weights
ReleasedAugust 13, 2025July 6, 2026
Context window131K262K
Input $/Mtok$0.40$0.0825
Output $/Mtok$2.00$0.33
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it01
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Mistral Medium 3.1 better than Hy3 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Mistral Medium 3.1 for work measured by SciCode or ALE-Bench, where it leads.

Which is cheaper, Mistral Medium 3.1 or Hy3?

Mistral Medium 3.1 costs $0.40 per million input tokens and $2.00 per million output tokens; Hy3 costs $0.0825 and $0.33. Cost per solved task is known for neither of them.

Which has the bigger context window, Mistral Medium 3.1 or Hy3?

Hy3, with 262K tokens against 131K.

How many benchmarks is the Mistral Medium 3.1 vs Hy3 comparison based on?

41 benchmarks have a result for both models as of October 21, 2026. Mistral Medium 3.1's coding index rests on 0 independent benchmarks and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next