Enter opens the highlighted result. Esc closes.

Command A vs DeepSeek V3.2

Updated October 21, 2026. 53 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Command A Replace DeepSeek V3.2
Command ACohere
Coding
No data
Overall
13.1
Output $/M
$10.00
3
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
21.5
Output $/M
$0.88
50
vs boards won of 53 shared

Decision reading

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

Command A has the edge where the context window is larger, 256K against 164K tokens, and where what matters is LegalBench (Vals AI) or CaseLaw (Vals AI), where it leads.

Model ACommand ACohereNot benchmarked1 independent benchmark$2.69 per solved taskModel BDeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved task
Shared results
53
Command A only
7
DeepSeek V3.2 only
22
Like-for-like categories
7 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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
Command A not benchmarked, DeepSeek V3.2 30.0 from 3 benchmarks.
Widest category
Math, 45.6 points apart on average over 6 shared benchmarks.
Widest single gap
AA Math Index: Command A 13.0 against DeepSeek V3.2 92.0, 79.0 points apart on a 0 to 100 scale.
Output price
Command A $10.00 against DeepSeek V3.2 $0.88 per million output tokens, 11.4x apart.
Context window
Command A takes 256K tokens, DeepSeek V3.2 164K.

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
CategoryCommand ABarsDeepSeek V3.2Ties
Coding07
Intelligence01
Agentic work01
Reasoning06
Math06
Knowledge29
General120
All shared boards350
CodingLike-for-likeSix of these boards feed the Vibe Coding Index07
Coding benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
SWE-bench VerifiedNot measured70.0%Only DeepSeek V3.2 measured
SWE-bench MultilingualNot measured59.0%Only DeepSeek V3.2 measured
LMArena WebDevNot measured1362Only DeepSeek V3.2 measured
AA Coding Index27.844.2DeepSeek V3.2
LiveCodeBench (Vals AI)35.1%80.7%DeepSeek V3.2
SWE-bench Verified (Vals AI)7.8%67.6%DeepSeek V3.2
Vibe Code Bench (Vals AI)Not measured5.1%Only DeepSeek V3.2 measured
Terminal-Bench 2.1 (Vals AI)17.6%Not measuredOnly Command A measured
LiveCodeBench (Artificial Analysis run)28.7%86.2%DeepSeek V3.2
SciCode38.5%Not measuredOnly Command A measured
Terminal-Bench Hard (Artificial Analysis run)25.0%35.6%DeepSeek V3.2
Terminal-BenchNot measured39.6%Only DeepSeek V3.2 measured
LMArena Text: coding13301453DeepSeek V3.2
LMArena Text: software and IT13361449DeepSeek V3.2
LMArena WebDev: ReactNot measured1350Only DeepSeek V3.2 measured
LMArena WebDev: HTMLNot measured1371Only DeepSeek V3.2 measured
Neither model has a result on the other 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
AA Intelligence Index13.121.5DeepSeek V3.2
Epoch Capabilities IndexNot measured146.1Only DeepSeek V3.2 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
AA Agentic Index1.0Not measuredOnly Command A measured
Finance Agent (Vals AI)9.0%Not measuredOnly Command A measured
τ²-Bench Telecom80.7%90.6%DeepSeek V3.2
APEX-AgentsNot measured21.3%Only DeepSeek V3.2 measured
Vending-Bench 2Not measured$1,034Only DeepSeek V3.2 measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
GPQA Diamond48.5%80.3%DeepSeek V3.2
GPQA Diamond (Artificial Analysis run)76.1%84.0%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)12.0%24.6%DeepSeek V3.2
ARC-AGINot measured57.0%Only DeepSeek V3.2 measured
ARC-AGI-2Not measured4.0%Only DeepSeek V3.2 measured
LMArena Text: hard prompts13261434DeepSeek V3.2
LMArena Text: expert12961436DeepSeek V3.2
LMArena Text: hard prompts in English13351446DeepSeek V3.2
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
AIME13.3%84.6%DeepSeek V3.2
MATH 500 (Vals AI)76.2%Not measuredOnly Command A measured
MGSM (Vals AI)85.7%86.0%DeepSeek V3.2
AA Math Index13.092.0DeepSeek V3.2
AIME 2025 (Artificial Analysis run)13.0%92.0%DeepSeek V3.2
LMArena Text: math13001435DeepSeek V3.2
LMArena Text: mathematical work13121440DeepSeek V3.2
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index29
Knowledge benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
MMLU Pro69.2%84.9%DeepSeek V3.2
LegalBench (Vals AI)79.7%76.1%Command A
CaseLaw (Vals AI)64.5%55.4%Command A
CorpFin (Vals AI)46.0%51.0%DeepSeek V3.2
TaxEval (Vals AI)61.4%68.2%DeepSeek V3.2
MedQA (Vals AI)80.5%93.9%DeepSeek V3.2
MedCode (Vals AI)19.7%Not measuredOnly Command A measured
MMLU Pro (Artificial Analysis run)71.2%86.2%DeepSeek V3.2
LMArena Text: science13391446DeepSeek V3.2
LMArena Text: medicine and healthcare13341440DeepSeek V3.2
LMArena Text: legal and government13481425DeepSeek V3.2
LMArena Text: business and finance13151410DeepSeek V3.2
Neither model has a result on the other 5 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
Design Arena: websitesNot measured1175Only DeepSeek V3.2 measured
Design Arena: UI componentsNot measured1149Only DeepSeek V3.2 measured
Design Arena: game devNot measured1131Only DeepSeek V3.2 measured
Design Arena: data vizNot measured1158Only DeepSeek V3.2 measured
Design Arena: 3DNot measured1133Only DeepSeek V3.2 measured
Design Arena: all code categoriesNot measured1164Only DeepSeek V3.2 measured
Design Arena: SVGNot measured1032Only DeepSeek V3.2 measured
Design Arena: ASCII artNot measured1082Only 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 index120
General benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
MedScribe (Vals AI)55.7%Not measuredOnly Command A measured
IFBench73.9%60.7%Command A
Long context reasoning (AA-LCR)52.7%73.3%DeepSeek V3.2
DTBench61.3%85.6%DeepSeek V3.2
LMCA10.3%28.8%DeepSeek V3.2
CL-benchNot measured12.4%Only DeepSeek V3.2 measured
CL-bench LifeNot measured7.4%Only DeepSeek V3.2 measured
LMArena Text13311425DeepSeek V3.2
LMArena Text: instruction following13091413DeepSeek V3.2
LMArena Text: creative writing13191400DeepSeek V3.2
LMArena Text: English13441439DeepSeek V3.2
LMArena Text: non-English13131409DeepSeek V3.2
LMArena Text: Chinese13271455DeepSeek V3.2
LMArena Text: German13421413DeepSeek V3.2
LMArena Text: Japanese12851374DeepSeek V3.2
LMArena Text: Korean12851370DeepSeek V3.2
LMArena Text: French13511429DeepSeek V3.2
LMArena Text: Spanish13471424DeepSeek V3.2
LMArena Text: Russian13141424DeepSeek V3.2
LMArena Text: Polish13241419DeepSeek V3.2
LMArena Text: multi-turn13381427DeepSeek V3.2
LMArena Text: longer queries13341428DeepSeek V3.2
LMArena Text: writing and language13281407DeepSeek V3.2
LMArena Text: entertainment and media13061395DeepSeek V3.2
Neither model has a result on the other 22 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Command A and DeepSeek V3.2
BenchmarkCommand ADeepSeek V3.2Leader
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

Command A$0.0075Fits in one request
DeepSeek V3.2$0.0007Fits in one request

DeepSeek V3.2 costs less

Repository review

50K input + 3K output tokens

Command A$0.155Fits in one request
DeepSeek V3.2$0.017Fits in one request

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

Command A$0.600Fits in one request
DeepSeek V3.2$0.067Exceeds 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 Command A and DeepSeek V3.2
SpecificationCohere Command ADeepSeek V3.2
MakerCohereDeepSeek
WeightsOpen weightsOpen weights
ReleasedMarch 13, 2025December 1, 2025
Context window256K164K
Input $/Mtok$2.50$0.29
Output $/Mtok$10.00$0.88
Coding indexNot benchmarked30.0
Independent measurements behind it13
Cost per solved task$2.69$0.99
SourceOpenRouter listingOpenRouter listing

Questions

Is Command A better than DeepSeek V3.2 for coding?

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; Command A has none, so the two cannot be ranked on coding. Command A has the edge where the context window is larger, 256K against 164K tokens, and where what matters is LegalBench (Vals AI) or CaseLaw (Vals AI), where it leads.

Which is cheaper, Command A or DeepSeek V3.2?

Command A costs $2.50 per million input tokens and $10.00 per million output tokens; DeepSeek V3.2 costs $0.29 and $0.88. Per solved SWE-bench Verified task, Command A costs $2.69 and DeepSeek V3.2 $0.99.

Which has the bigger context window, Command A or DeepSeek V3.2?

Command A, with 256K tokens against 164K.

How many benchmarks is the Command A vs DeepSeek V3.2 comparison based on?

53 benchmarks have a result for both models as of October 21, 2026. Command A's coding index rests on 1 independent benchmark and DeepSeek V3.2's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next