Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs Gemini 2.5 Flash

DeepSeek V3.2 leads Gemini 2.5 Flash on the Artificial Analysis Intelligence Index, 21.5 to 15.5, and DeepSeek V3.2 costs 2.8x less per output token ($0.88 vs $2.50 per million). Updated October 21, 2026; 54 benchmarks have a result for both.

Replace DeepSeek V3.2 Replace Gemini 2.5 Flash
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
37.9
Output $/M
$0.88
47
Gemini 2.5 FlashGoogle
Coding
No data
Overall
34.3
Output $/M
$2.50
7
vs boards won of 54 shared

DeepSeek V3.2 or Gemini 2.5 Flash: which to pick

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; Gemini 2.5 Flash has none, so the two cannot be ranked on coding.

Pick Gemini 2.5 Flash for a lower cost per solved task ($0.466 against $0.99) and a larger context window (1.05M against 164K tokens).

Model ADeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved taskModel BGemini 2.5 FlashGoogleNot benchmarkedNo coding benchmark$0.466 per solved task
Shared results
54
DeepSeek V3.2 only
21
Gemini 2.5 Flash only
19
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, Gemini 2.5 Flash not benchmarked.
Widest category
Agentic work, 31.0 points apart on average over 2 shared benchmarks.
Widest single gap
τ²-Bench Telecom: DeepSeek V3.2 90.6% against Gemini 2.5 Flash 31.6%, 59.1 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.88 against Gemini 2.5 Flash $2.50 per million output tokens, 2.8x apart.
Context window
DeepSeek V3.2 takes 164K tokens, Gemini 2.5 Flash 1.05M.

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 V3.2BarsGemini 2.5 FlashTies
Coding60
Intelligence20
Agentic work20
Reasoning70
Math40
Knowledge41
Web design70
General156
All shared boards477
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
Coding benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
SWE-bench Verified70.0%28.7%DeepSeek V3.2
SWE-bench Multilingual59.0%Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev1362Not measuredOnly DeepSeek V3.2 measured
AA Coding Index44.2Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Vals AI)80.7%Not measuredOnly DeepSeek V3.2 measured
SWE-bench Verified (Vals AI)67.6%Not measuredOnly DeepSeek V3.2 measured
Vibe Code Bench (Vals AI)5.1%Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Artificial Analysis run)86.2%69.5%DeepSeek V3.2
Terminal-Bench Hard (Artificial Analysis run)35.6%13.6%DeepSeek V3.2
Terminal-Bench39.6%17.1%DeepSeek V3.2
LMArena Text: coding14531423DeepSeek V3.2
LMArena Text: software and IT14491425DeepSeek V3.2
LMArena WebDev: React1350Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev: HTML1371Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 25 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
AA Intelligence Index21.515.5DeepSeek V3.2
Epoch Capabilities Index146.1142.9DeepSeek V3.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
Agentic work benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
τ²-Bench Telecom90.6%31.6%DeepSeek V3.2
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
BALROGNot measured33.5%Only Gemini 2.5 Flash measured
Vending-Bench 2$1,034$549DeepSeek V3.2
Berkeley Function Calling Leaderboard (BFCL)Not measured56.2%Only Gemini 2.5 Flash measured
BrowseComp-PlusNot measured39.9%Only Gemini 2.5 Flash measured
Neither model has a result on the other 35 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index70
Reasoning benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
GPQA Diamond80.3%Not measuredOnly DeepSeek V3.2 measured
GPQA Diamond (Artificial Analysis run)84.0%79.0%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)24.6%12.1%DeepSeek V3.2
ARC-AGI57.0%33.3%DeepSeek V3.2
ARC-AGI-24.0%2.5%DeepSeek V3.2
SimpleBenchNot measured41.2%Only Gemini 2.5 Flash measured
LMArena Text: hard prompts14341422DeepSeek V3.2
LMArena Text: expert14361426DeepSeek V3.2
LMArena Text: hard prompts in English14461420DeepSeek V3.2
LMArena Vision: diagramsNot measured1245Only Gemini 2.5 Flash measured
LMArena Vision: homeworkNot measured1244Only Gemini 2.5 Flash measured
Kagi LLM BenchmarkNot measured56.8%Only Gemini 2.5 Flash measured
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
AIME84.6%Not measuredOnly DeepSeek V3.2 measured
MGSM (Vals AI)86.0%Not measuredOnly DeepSeek V3.2 measured
AA Math Index92.073.3DeepSeek V3.2
AIME 2025 (Artificial Analysis run)92.0%73.3%DeepSeek V3.2
LMArena Text: math14351411DeepSeek V3.2
LMArena Text: mathematical work14401422DeepSeek V3.2
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index41
Knowledge benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
MMLU Pro84.9%Not measuredOnly DeepSeek V3.2 measured
LegalBench (Vals AI)76.1%Not measuredOnly DeepSeek V3.2 measured
CaseLaw (Vals AI)55.4%Not measuredOnly DeepSeek V3.2 measured
CorpFin (Vals AI)51.0%Not measuredOnly DeepSeek V3.2 measured
TaxEval (Vals AI)68.2%Not measuredOnly DeepSeek V3.2 measured
MedQA (Vals AI)93.9%Not measuredOnly DeepSeek V3.2 measured
MedCode (Vals AI)Not measured40.4%Only Gemini 2.5 Flash measured
MMLU Pro (Artificial Analysis run)86.2%83.2%DeepSeek V3.2
LMArena Text: science14461433DeepSeek V3.2
LMArena Text: medicine and healthcare14401428DeepSeek V3.2
LMArena Text: legal and government14251430Gemini 2.5 Flash
LMArena Text: business and finance14101404DeepSeek V3.2
FACTS Benchmark SuiteNot measured47.2%Only Gemini 2.5 Flash measured
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index70
Web design benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
Design Arena: websites11751074DeepSeek V3.2
Design Arena: UI components11491005DeepSeek V3.2
Design Arena: game dev11321005DeepSeek V3.2
Design Arena: data viz11581060DeepSeek V3.2
Design Arena: 3D11331021DeepSeek V3.2
Design Arena: all code categories11641059DeepSeek V3.2
Design Arena: SVG10321020DeepSeek V3.2
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 index156
General benchmarks for DeepSeek V3.2 and Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
SAGE (Vals AI)Not measured44.8%Only Gemini 2.5 Flash measured
MedScribe (Vals AI)Not measured83.0%Only Gemini 2.5 Flash measured
IFBench60.7%50.3%DeepSeek V3.2
Long context reasoning (AA-LCR)73.3%65.3%DeepSeek V3.2
DTBench85.6%76.5%DeepSeek V3.2
LMCA28.8%27.5%DeepSeek V3.2
CL-bench12.4%Not measuredOnly DeepSeek V3.2 measured
CL-bench Life7.4%Not measuredOnly DeepSeek V3.2 measured
LMArena Text14251417DeepSeek V3.2
LMArena Text: instruction following14131405DeepSeek V3.2
LMArena Text: creative writing14001400Gemini 2.5 Flash
LMArena VisionNot measured1235Only Gemini 2.5 Flash measured
LMArena Text: English14391418DeepSeek V3.2
LMArena Text: non-English14091409Gemini 2.5 Flash
LMArena Text: Chinese14551449DeepSeek V3.2
LMArena Text: German14131417Gemini 2.5 Flash
LMArena Text: Japanese13741396Gemini 2.5 Flash
LMArena Text: Korean13701385Gemini 2.5 Flash
LMArena Text: French14291432Gemini 2.5 Flash
LMArena Text: Spanish14241422DeepSeek V3.2
LMArena Text: Russian14241415DeepSeek V3.2
LMArena Text: Polish14191413DeepSeek V3.2
LMArena Text: multi-turn14271408DeepSeek V3.2
LMArena Text: longer queries14281419DeepSeek V3.2
LMArena Text: writing and language14071405DeepSeek V3.2
LMArena Text: entertainment and media13951393DeepSeek V3.2
LMArena Vision: EnglishNot measured1239Only Gemini 2.5 Flash measured
LMArena Vision: ChineseNot measured1284Only Gemini 2.5 Flash measured
LMArena Vision: reading textNot measured1237Only Gemini 2.5 Flash measured
LMArena Vision: humorNot measured1225Only Gemini 2.5 Flash measured
LMArena Vision: captioningNot measured1277Only Gemini 2.5 Flash measured
LMArena Vision: recognizing thingsNot measured1252Only Gemini 2.5 Flash measured
LMArena Vision: creative writingNot measured1241Only Gemini 2.5 Flash measured
Neither model has a result on the other 13 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 Gemini 2.5 Flash
BenchmarkDeepSeek V3.2Gemini 2.5 FlashLeader
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
Gemini 2.5 Flash$0.0015Fits in one request

DeepSeek V3.2 costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.017Fits in one request
Gemini 2.5 Flash$0.022Fits in one request

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.067Exceeds the context window
Gemini 2.5 Flash$0.085Fits in one request

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 Gemini 2.5 Flash
SpecificationDeepSeek V3.2Google Gemini 2.5 Flash
MakerDeepSeekGoogle
WeightsOpen weightsNo public weights listed
ReleasedDecember 1, 2025June 17, 2025
Context window164K1.05M
Input $/Mtok$0.29$0.30
Output $/Mtok$0.88$2.50
Coding index30.0Not benchmarked
Independent measurements behind it30
Cost per solved task$0.99$0.466
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than Gemini 2.5 Flash for coding?

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; Gemini 2.5 Flash has none, so the two cannot be ranked on coding. Pick Gemini 2.5 Flash for a lower cost per solved task ($0.466 against $0.99) and a larger context window (1.05M against 164K tokens).

Which is cheaper, DeepSeek V3.2 or Gemini 2.5 Flash?

DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; Gemini 2.5 Flash costs $0.30 and $2.50. Per solved SWE-bench Verified task, DeepSeek V3.2 costs $0.99 and Gemini 2.5 Flash $0.466.

Which has the bigger context window, DeepSeek V3.2 or Gemini 2.5 Flash?

Gemini 2.5 Flash, with 1.05M tokens against 164K.

How many benchmarks is the DeepSeek V3.2 vs Gemini 2.5 Flash comparison based on?

54 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and Gemini 2.5 Flash's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next