Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs MiniMax M2.5

MiniMax M2.5 leads DeepSeek V3.2 on coding, 37.2 to 30.0 on the Vibe Coding Index, and DeepSeek V3.2 costs 1.4x less per output token ($0.88 vs $1.20 per million). Updated October 22, 2026; 65 benchmarks have a result for both.

Replace DeepSeek V3.2 Replace MiniMax M2.5
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
37.9
Output $/M
$0.88
37
MiniMax M2.5MiniMax
Coding
37.2
Overall
37.3
Output $/M
$1.20
26
vs boards won of 65 shared, 2 tied

DeepSeek V3.2 or MiniMax M2.5: which to pick

Pick MiniMax M2.5 for coding: it scores 37.2 on the Vibe Coding Index against 30.0 for DeepSeek V3.2.

Pick DeepSeek V3.2 for output tokens 1.4x cheaper ($0.88 vs $1.20 per million) and work measured by LiveCodeBench (Vals AI) or Terminal-Bench Hard (Artificial Analysis run), where it leads.

Model ADeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved taskModel BMiniMax M2.5MiniMax37.2/1003 independent benchmarks$0.62 per solved task
Shared results
65
DeepSeek V3.2 only
10
MiniMax M2.5 only
3
Like-for-like categories
8 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
DeepSeek V3.2 30.0 from 3 benchmarks, MiniMax M2.5 37.2 from 3 benchmarks.
Widest category
General, 11.1 points apart on average over 21 shared benchmarks.
Widest single gap
LMArena Text: Japanese: DeepSeek V3.2 1374 against MiniMax M2.5 1171, 33.8 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.88 against MiniMax M2.5 $1.20 per million output tokens, 1.4x apart.
Context window
DeepSeek V3.2 takes 164K tokens, MiniMax M2.5 205K.

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
CategoryDeepSeek V3.2BarsMiniMax M2.5Ties
Coding48
Intelligence02
Agentic work11
Reasoning44
Math21
Knowledge721 tie
Web design07
General1911 tie
All shared boards37262 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index48
Coding benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
SWE-bench Verified70.0%75.8%MiniMax M2.5
SWE-bench Multilingual59.0%68.3%MiniMax M2.5
LMArena WebDev13621386MiniMax M2.5
AA Coding Index44.2Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Vals AI)80.7%79.2%DeepSeek V3.2
SWE-bench Verified (Vals AI)67.6%74.2%MiniMax M2.5
Vibe Code Bench (Vals AI)5.1%14.8%MiniMax M2.5
LiveCodeBench (Artificial Analysis run)86.2%Not measuredOnly DeepSeek V3.2 measured
Terminal-Bench Hard (Artificial Analysis run)35.6%34.8%DeepSeek V3.2
Terminal-Bench39.6%42.7%MiniMax M2.5
ALE-BenchNot measured618.2Only MiniMax M2.5 measured
LMArena Text: coding14531380DeepSeek V3.2
LMArena Text: software and IT14491383DeepSeek V3.2
LMArena WebDev: React13501374MiniMax M2.5
LMArena WebDev: HTML13711414MiniMax M2.5
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
AA Intelligence Index21.522.8MiniMax M2.5
Epoch Capabilities Index146.1146.6MiniMax M2.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
τ²-Bench Telecom90.6%95.3%MiniMax M2.5
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
Vending-Bench 2$1,034-$23DeepSeek V3.2
Neither model has a result on the other 38 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index44
Reasoning benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
GPQA Diamond80.3%82.1%MiniMax M2.5
GPQA Diamond (Artificial Analysis run)84.0%84.8%MiniMax M2.5
Humanity's Last Exam (Artificial Analysis run)24.6%20.5%DeepSeek V3.2
ARC-AGI57.0%63.7%MiniMax M2.5
ARC-AGI-24.0%4.9%MiniMax M2.5
LMArena Text: hard prompts14341372DeepSeek V3.2
LMArena Text: expert14361378DeepSeek V3.2
LMArena Text: hard prompts in English14461384DeepSeek V3.2
Kagi LLM BenchmarkNot measured55.2%Only MiniMax M2.5 measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index21
Math benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
AIME84.6%88.8%MiniMax M2.5
MGSM (Vals AI)86.0%Not measuredOnly DeepSeek V3.2 measured
AA Math Index92.0Not measuredOnly DeepSeek V3.2 measured
AIME 2025 (Artificial Analysis run)92.0%Not measuredOnly DeepSeek V3.2 measured
ProofBenchNot measured4.0%Only MiniMax M2.5 measured
LMArena Text: math14351378DeepSeek V3.2
LMArena Text: mathematical work14401378DeepSeek V3.2
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index72
Knowledge benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
MMLU Pro84.9%80.1%DeepSeek V3.2
LegalBench (Vals AI)76.1%80.0%MiniMax M2.5
CaseLaw (Vals AI)55.4%53.5%DeepSeek V3.2
CorpFin (Vals AI)51.0%59.6%MiniMax M2.5
TaxEval (Vals AI)68.2%68.2%Tie
MedQA (Vals AI)93.9%92.5%DeepSeek V3.2
MMLU Pro (Artificial Analysis run)86.2%Not measuredOnly DeepSeek V3.2 measured
LMArena Text: science14461369DeepSeek V3.2
LMArena Text: medicine and healthcare14401357DeepSeek V3.2
LMArena Text: legal and government14251366DeepSeek V3.2
LMArena Text: business and finance14101356DeepSeek V3.2
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
Design Arena: websites11751222MiniMax M2.5
Design Arena: UI components11461173MiniMax M2.5
Design Arena: game dev11311176MiniMax M2.5
Design Arena: data viz11581172MiniMax M2.5
Design Arena: 3D11321170MiniMax M2.5
Design Arena: all code categories11631207MiniMax M2.5
Design Arena: SVG10321150MiniMax M2.5
Design Arena: ASCII art1083Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index191
General benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
IFBench60.7%71.6%MiniMax M2.5
Long context reasoning (AA-LCR)73.3%73.3%Tie
DTBench85.6%Not measuredOnly DeepSeek V3.2 measured
LMCA28.8%Not measuredOnly DeepSeek V3.2 measured
CL-bench12.4%11.4%DeepSeek V3.2
CL-bench Life7.4%6.3%DeepSeek V3.2
LMArena Text14251359DeepSeek V3.2
LMArena Text: instruction following14131353DeepSeek V3.2
LMArena Text: creative writing14001331DeepSeek V3.2
LMArena Text: English14391380DeepSeek V3.2
LMArena Text: non-English14091338DeepSeek V3.2
LMArena Text: Chinese14551393DeepSeek V3.2
LMArena Text: German14131361DeepSeek V3.2
LMArena Text: Japanese13741171DeepSeek V3.2
LMArena Text: Korean13701233DeepSeek V3.2
LMArena Text: French14291363DeepSeek V3.2
LMArena Text: Spanish14241353DeepSeek V3.2
LMArena Text: Russian14241359DeepSeek V3.2
LMArena Text: Polish14191334DeepSeek V3.2
LMArena Text: multi-turn14271364DeepSeek V3.2
LMArena Text: longer queries14281366DeepSeek V3.2
LMArena Text: writing and language14071345DeepSeek V3.2
LMArena Text: entertainment and media13951329DeepSeek V3.2
Neither model has a result on the other 23 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V3.2 and MiniMax M2.5
BenchmarkDeepSeek V3.2MiniMax M2.5Leader
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

DeepSeek V3.2$0.0007Fits in one request
MiniMax M2.5$0.0009Fits in one request

DeepSeek V3.2 costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.017Fits in one request
MiniMax M2.5$0.019Fits in one request

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.067Exceeds the context window
MiniMax M2.5$0.072Exceeds the context window

DeepSeek V3.2 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 DeepSeek V3.2 and MiniMax M2.5
SpecificationDeepSeek V3.2MiniMax M2.5
MakerDeepSeekMiniMax
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025February 12, 2026
Context window164K205K
Input $/Mtok$0.29$0.30
Output $/Mtok$0.88$1.20
Coding index30.037.2
Independent measurements behind it33
Cost per solved task$0.99$0.62
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than MiniMax M2.5 for coding?

Pick MiniMax M2.5 for coding: it scores 37.2 on the Vibe Coding Index against 30.0 for DeepSeek V3.2. Pick DeepSeek V3.2 for output tokens 1.4x cheaper ($0.88 vs $1.20 per million) and work measured by LiveCodeBench (Vals AI) or Terminal-Bench Hard (Artificial Analysis run), where it leads.

Which is cheaper, DeepSeek V3.2 or MiniMax M2.5?

DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; MiniMax M2.5 costs $0.30 and $1.20. Per solved SWE-bench Verified task, DeepSeek V3.2 costs $0.99 and MiniMax M2.5 $0.62.

Which has the bigger context window, DeepSeek V3.2 or MiniMax M2.5?

MiniMax M2.5, with 205K tokens against 164K.

How many benchmarks is the DeepSeek V3.2 vs MiniMax M2.5 comparison based on?

65 benchmarks have a result for both models as of October 22, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and MiniMax M2.5's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v3-2-vs-minimax-minimax-m2-5" title="DeepSeek V3.2 vs MiniMax M2.5, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v3-2-vs-minimax-minimax-m2-5">LLMBenchmark.ai, DeepSeek V3.2 vs MiniMax M2.5</a></p>