Enter opens the highlighted result. Esc closes.

MiniMax M2.1 vs Hy3

Hy3 costs 2.3x less per output token than MiniMax M2.1 ($0.528 vs $1.20 per million). Updated October 22, 2026; 11 benchmarks have a result for both.

Replace MiniMax M2.1 Replace Hy3
MiniMax M2.1MiniMax
Coding
No data
Overall
No data
Output $/M
$1.20
2
Hy3Tencent
Coding
No data
Overall
No data
Output $/M
$0.528
9
vs boards won of 11 shared

MiniMax M2.1 or Hy3: which to pick

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

Pick MiniMax M2.1 for work measured by ALE-Bench or GPQA Diamond (Artificial Analysis run), where it leads.

Model AMiniMax M2.1MiniMaxNot benchmarked1 independent benchmark$0.59 per solved taskModel BHy3TencentNot benchmarked1 independent benchmarkNo cost per solved task
Shared results
11
MiniMax M2.1 only
23
Hy3 only
33
Like-for-like categories
4 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
MiniMax M2.1 not benchmarked, Hy3 not benchmarked.
Widest category
General, 11.3 points apart on average over 1 shared benchmark.
Widest single gap
Long context reasoning (AA-LCR): MiniMax M2.1 67.7% against Hy3 79.0%, 11.3 points apart on a 0 to 100 scale.
Output price
MiniMax M2.1 $1.20 against Hy3 $0.528 per million output tokens, 2.3x apart.
Context window
MiniMax M2.1 takes 205K 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
CategoryMiniMax M2.1BarsHy3Ties
Coding01
Reasoning02
Web design25
General01
All shared boards29
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
Coding benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
LMArena WebDevNot measured1507Only Hy3 measured
LiveCodeBench (Vals AI)81.8%Not measuredOnly MiniMax M2.1 measured
SWE-bench Verified (Vals AI)74.8%Not measuredOnly MiniMax M2.1 measured
LiveCodeBench (Artificial Analysis run)81.0%Not measuredOnly MiniMax M2.1 measured
SciCodeNot measured48.6%Only Hy3 measured
Terminal-Bench Hard (Artificial Analysis run)28.8%Not measuredOnly MiniMax M2.1 measured
Terminal-Bench36.6%Not measuredOnly MiniMax M2.1 measured
ALE-Bench623.8838.6Hy3
LMArena Text: codingNot measured1466Only Hy3 measured
LMArena Text: software and ITNot measured1463Only Hy3 measured
LMArena WebDev: ReactNot measured1511Only Hy3 measured
LMArena WebDev: HTMLNot measured1487Only Hy3 measured
Neither model has a result on the other 27 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Intelligence benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
AA Intelligence Index20.9Not measuredOnly MiniMax M2.1 measured
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Agentic work benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
τ²-Bench Telecom85.4%Not measuredOnly MiniMax M2.1 measured
Neither model has a result on the other 40 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
Reasoning benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
GPQA Diamond78.5%Not measuredOnly MiniMax M2.1 measured
GPQA Diamond (Artificial Analysis run)83.0%89.7%Hy3
Humanity's Last Exam (Artificial Analysis run)23.2%33.5%Hy3
LMArena Text: hard promptsNot measured1447Only Hy3 measured
LMArena Text: expertNot measured1461Only Hy3 measured
LMArena Text: hard prompts in EnglishNot measured1455Only Hy3 measured
Neither model has a result on the other 16 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Math benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
AIME77.9%Not measuredOnly MiniMax M2.1 measured
MATH 500 (Vals AI)89.0%Not measuredOnly MiniMax M2.1 measured
MGSM (Vals AI)87.8%Not measuredOnly MiniMax M2.1 measured
AA Math Index82.7Not measuredOnly MiniMax M2.1 measured
AIME 2025 (Artificial Analysis run)82.7%Not measuredOnly MiniMax M2.1 measured
LMArena Text: mathNot measured1473Only Hy3 measured
LMArena Text: mathematical workNot measured1482Only Hy3 measured
Neither model has a result on the other 6 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
Knowledge benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
MMLU Pro87.0%Not measuredOnly MiniMax M2.1 measured
LegalBench (Vals AI)75.5%Not measuredOnly MiniMax M2.1 measured
CaseLaw (Vals AI)55.8%Not measuredOnly MiniMax M2.1 measured
CorpFin (Vals AI)58.9%Not measuredOnly MiniMax M2.1 measured
TaxEval (Vals AI)66.3%Not measuredOnly MiniMax M2.1 measured
MedQA (Vals AI)91.2%Not measuredOnly MiniMax M2.1 measured
MedCode (Vals AI)34.1%Not measuredOnly MiniMax M2.1 measured
MMLU Pro (Artificial Analysis run)87.5%Not measuredOnly MiniMax M2.1 measured
LMArena Text: scienceNot measured1457Only Hy3 measured
LMArena Text: medicine and healthcareNot measured1452Only Hy3 measured
LMArena Text: legal and governmentNot measured1422Only Hy3 measured
LMArena Text: business and financeNot measured1424Only Hy3 measured
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index25
Web design benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
Design Arena: websites12011230Hy3
Design Arena: UI components12201204MiniMax M2.1
Design Arena: game dev11341208Hy3
Design Arena: data viz12061164MiniMax M2.1
Design Arena: 3D11641232Hy3
Design Arena: all code categories11901227Hy3
Design Arena: SVG11311200Hy3
Design Arena: ASCII artNot measured1152Only Hy3 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
General benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
MedScribe (Vals AI)80.8%Not measuredOnly MiniMax M2.1 measured
IFBench69.9%Not measuredOnly MiniMax M2.1 measured
Long context reasoning (AA-LCR)67.7%79.0%Hy3
LMArena TextNot measured1438Only Hy3 measured
LMArena Text: instruction followingNot measured1426Only Hy3 measured
LMArena Text: creative writingNot measured1403Only Hy3 measured
LMArena Text: EnglishNot measured1448Only Hy3 measured
LMArena Text: non-EnglishNot measured1426Only Hy3 measured
LMArena Text: ChineseNot measured1485Only Hy3 measured
LMArena Text: GermanNot measured1449Only Hy3 measured
LMArena Text: JapaneseNot measured1397Only Hy3 measured
LMArena Text: KoreanNot measured1396Only Hy3 measured
LMArena Text: FrenchNot measured1456Only Hy3 measured
LMArena Text: SpanishNot measured1457Only Hy3 measured
LMArena Text: RussianNot measured1432Only Hy3 measured
LMArena Text: PolishNot measured1447Only Hy3 measured
LMArena Text: multi-turnNot measured1436Only Hy3 measured
LMArena Text: longer queriesNot measured1442Only Hy3 measured
LMArena Text: writing and languageNot measured1414Only Hy3 measured
LMArena Text: entertainment and mediaNot measured1400Only Hy3 measured
Neither model has a result on the other 26 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for MiniMax M2.1 and Hy3
BenchmarkMiniMax M2.1Hy3Leader
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

MiniMax M2.1$0.0009Fits in one request
Hy3$0.0004Fits in one request

Hy3 costs less

Repository review

50K input + 3K output tokens

MiniMax M2.1$0.019Fits in one request
Hy3$0.0082Fits in one request

Hy3 costs less

Agent loop

200K input + 10K output tokens

MiniMax M2.1$0.072Exceeds the context window
Hy3$0.032Fits 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 MiniMax M2.1 and Hy3
SpecificationMiniMax M2.1Tencent Hy3
MakerMiniMaxTencent
WeightsOpen weightsOpen weights
ReleasedDecember 23, 2025July 6, 2026
Context window205K262K
Input $/Mtok$0.30$0.132
Output $/Mtok$1.20$0.528
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it11
Cost per solved task$0.59Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is MiniMax M2.1 better than Hy3 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick MiniMax M2.1 for work measured by ALE-Bench or GPQA Diamond (Artificial Analysis run), where it leads.

Which is cheaper, MiniMax M2.1 or Hy3?

MiniMax M2.1 costs $0.30 per million input tokens and $1.20 per million output tokens; Hy3 costs $0.132 and $0.528. Cost per solved task is known for MiniMax M2.1 only.

Which has the bigger context window, MiniMax M2.1 or Hy3?

Hy3, with 262K tokens against 205K.

How many benchmarks is the MiniMax M2.1 vs Hy3 comparison based on?

11 benchmarks have a result for both models as of October 22, 2026. MiniMax M2.1's coding index rests on 1 independent benchmark and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next