Enter opens the highlighted result. Esc closes.

UI-TARS 7B vs Command A

UI-TARS 7B costs 50.0x less per output token than Command A ($0.20 vs $10.00 per million). Updated October 21, 2026; 1 benchmarks have a result for both.

Replace UI-TARS 7B Replace Command A
UI-TARS 7BByteDance
Coding
No data
Overall
No data
Output $/M
$0.20
0
Command ACohere
Coding
No data
Overall
No data
Output $/M
$10.00
1
vs boards won of 1 shared

UI-TARS 7B or Command A: which to pick

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

Pick UI-TARS 7B for output tokens 50.0x cheaper ($0.20 vs $10.00 per million) and work measured by Kagi LLM Benchmark, where it leads.

Model AUI-TARS 7BByteDanceNot benchmarkedNo coding benchmarkNo cost per solved taskModel BCommand ACohereNot benchmarked1 independent benchmark$2.69 per solved task
Shared results
1
UI-TARS 7B only
0
Command A only
61
Like-for-like categories
1 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
UI-TARS 7B not benchmarked, Command A not benchmarked.
Widest category
Reasoning, 8.1 points apart on average over 1 shared benchmark.
Widest single gap
Kagi LLM Benchmark: UI-TARS 7B 20.7% against Command A 28.8%, 8.1 points apart on a 0 to 100 scale.
Output price
UI-TARS 7B $0.20 against Command A $10.00 per million output tokens, 50.0x apart.
Context window
UI-TARS 7B takes 128K tokens, Command A 256K.

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
CategoryUI-TARS 7BBarsCommand ATies
Reasoning01
All shared boards01
CodingDirectional onlySix of these boards feed the Vibe Coding Index9 boards, none shared
Coding benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
AA Coding IndexNot measured27.8Only Command A measured
LiveCodeBench (Vals AI)Not measured35.1%Only Command A measured
SWE-bench Verified (Vals AI)Not measured7.8%Only Command A measured
Terminal-Bench 2.1 (Vals AI)Not measured17.6%Only Command A measured
LiveCodeBench (Artificial Analysis run)Not measured28.7%Only Command A measured
SciCodeNot measured38.5%Only Command A measured
Terminal-Bench Hard (Artificial Analysis run)Not measured25.0%Only Command A measured
LMArena Text: codingNot measured1330Only Command A measured
LMArena Text: software and ITNot measured1336Only Command A measured
Neither model has a result on the other 30 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
Intelligence benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
AA Intelligence IndexNot measured13.1Only Command A measured
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Agentic work benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
AA Agentic IndexNot measured1.0Only Command A measured
Finance Agent (Vals AI)Not measured9.0%Only Command A measured
τ²-Bench TelecomNot measured80.7%Only Command A measured
Berkeley Function Calling Leaderboard (BFCL)Not measured46.5%Only Command A measured
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index01
Reasoning benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
GPQA DiamondNot measured48.5%Only Command A measured
GPQA Diamond (Artificial Analysis run)Not measured76.1%Only Command A measured
Humanity's Last Exam (Artificial Analysis run)Not measured12.0%Only Command A measured
LMArena Text: hard promptsNot measured1326Only Command A measured
LMArena Text: expertNot measured1296Only Command A measured
LMArena Text: hard prompts in EnglishNot measured1335Only Command A measured
Kagi LLM Benchmark20.7%28.8%Command A
Neither model has a result on the other 15 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Math benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
AIMENot measured13.3%Only Command A measured
MATH 500 (Vals AI)Not measured76.2%Only Command A measured
MGSM (Vals AI)Not measured85.7%Only Command A measured
AA Math IndexNot measured13.0Only Command A measured
AIME 2025 (Artificial Analysis run)Not measured13.0%Only Command A measured
LMArena Text: mathNot measured1300Only Command A measured
LMArena Text: mathematical workNot measured1312Only Command A 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 UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
MMLU ProNot measured69.2%Only Command A measured
LegalBench (Vals AI)Not measured79.7%Only Command A measured
CaseLaw (Vals AI)Not measured64.5%Only Command A measured
CorpFin (Vals AI)Not measured46.0%Only Command A measured
TaxEval (Vals AI)Not measured61.4%Only Command A measured
MedQA (Vals AI)Not measured80.5%Only Command A measured
MedCode (Vals AI)Not measured19.7%Only Command A measured
MMLU Pro (Artificial Analysis run)Not measured71.2%Only Command A measured
LMArena Text: scienceNot measured1339Only Command A measured
LMArena Text: medicine and healthcareNot measured1334Only Command A measured
LMArena Text: legal and governmentNot measured1348Only Command A measured
LMArena Text: business and financeNot measured1315Only Command A measured
Neither model has a result on the other 6 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
Neither model has a result on 9 boards here
GeneralDirectional onlyPublic boards, shown as published, kept out of the index22 boards, none shared
General benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
MedScribe (Vals AI)Not measured55.7%Only Command A measured
IFBenchNot measured73.9%Only Command A measured
Long context reasoning (AA-LCR)Not measured52.7%Only Command A measured
DTBenchNot measured61.3%Only Command A measured
LMCANot measured10.3%Only Command A measured
LMArena TextNot measured1331Only Command A measured
LMArena Text: instruction followingNot measured1309Only Command A measured
LMArena Text: creative writingNot measured1319Only Command A measured
LMArena Text: EnglishNot measured1344Only Command A measured
LMArena Text: non-EnglishNot measured1313Only Command A measured
LMArena Text: ChineseNot measured1327Only Command A measured
LMArena Text: GermanNot measured1342Only Command A measured
LMArena Text: JapaneseNot measured1285Only Command A measured
LMArena Text: KoreanNot measured1285Only Command A measured
LMArena Text: FrenchNot measured1351Only Command A measured
LMArena Text: SpanishNot measured1347Only Command A measured
LMArena Text: RussianNot measured1314Only Command A measured
LMArena Text: PolishNot measured1324Only Command A measured
LMArena Text: multi-turnNot measured1338Only Command A measured
LMArena Text: longer queriesNot measured1334Only Command A measured
LMArena Text: writing and languageNot measured1328Only Command A measured
LMArena Text: entertainment and mediaNot measured1306Only Command A measured
Neither model has a result on the other 24 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for UI-TARS 7B and Command A
BenchmarkUI-TARS 7BCommand ALeader
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

UI-TARS 7B$0.0002Fits in one request
Command A$0.0075Fits in one request

UI-TARS 7B costs less

Repository review

50K input + 3K output tokens

UI-TARS 7B$0.0056Fits in one request
Command A$0.155Fits in one request

UI-TARS 7B costs less

Agent loop

200K input + 10K output tokens

UI-TARS 7B$0.022Exceeds the context window
Command A$0.600Fits in one request

UI-TARS 7B 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 UI-TARS 7B and Command A
SpecificationByteDance UI-TARS 7BCohere Command A
MakerByteDanceCohere
WeightsOpen weightsOpen weights
ReleasedJuly 22, 2025March 13, 2025
Context window128K256K
Input $/Mtok$0.10$2.50
Output $/Mtok$0.20$10.00
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it01
Cost per solved taskNot computable$2.69
SourceOpenRouter listingOpenRouter listing

Questions

Is UI-TARS 7B better than Command A for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick UI-TARS 7B for output tokens 50.0x cheaper ($0.20 vs $10.00 per million) and work measured by Kagi LLM Benchmark, where it leads.

Which is cheaper, UI-TARS 7B or Command A?

UI-TARS 7B costs $0.10 per million input tokens and $0.20 per million output tokens; Command A costs $2.50 and $10.00. Cost per solved task is known for Command A only.

Which has the bigger context window, UI-TARS 7B or Command A?

Command A, with 256K tokens against 128K.

How many benchmarks is the UI-TARS 7B vs Command A comparison based on?

1 benchmarks have a result for both models as of October 21, 2026. UI-TARS 7B's coding index rests on 0 independent benchmarks and Command A's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next