GPQA DiamondVerified
52.7%
GPQA Diamond (AA run)
Models / Cohere
Command A
Command · Released 2026-03-15
Cohere's enterprise-focused Command tier. Strong at RAG, tool use, and multilingual retrieval workflows. 256K context.
Pick when your stack already uses Cohere embed + rerank and you want one vendor for retrieval pipelines.
Composite20
Context256K
Input / 1M$2.5
Output / 1M$10
Knowledge cutoff2025-06-01
Output speed60 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
25/100 avg
- Reasoning
- 53
- Agentic
- 1
- Long context
- 21
3 tested benchmark families
Benchmark placements
Per-benchmark positions
Where Command A places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| AA output speed | Performance | #104/ 195 | Artificial Analysis | 2026-10-12 | |
| Terminal-Bench | Agentic | #164/ 167 | 1 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #176/ 184 | 53 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #178/ 184 | 21 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #184/ 186 | 4 | Artificial Analysis | 2026-09-28 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Command A shows up, it's highlighted.
ReasoningCommand A · 53
- 196
- 295
- 395
Math
- 152
- 252
- 350
AgenticCommand A · 1
- 191
- 289
- 389
Long contextCommand A · 21
- 189
- 286
- 385
Newest receipts
5 recentLast 120 days
Benchmark rows added to the public ledger for Command A in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 60 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 21.33% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 52.7% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 4% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 0.76% | Artificial Analysis | 24 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Command A sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Command A. Scores link to the original source; gaps mean no public row exists yet.
5 of 13 catalog benchmarks have a sourced row for Command A.
Reasoning
Humanity's Last ExamVerified
4%
HLE (AA run)
Agentic
Terminal-BenchVerified
0.76%
Terminal-Bench Hard (AA run)
Long context
AA-LCRVerified
21.33%
AA-LCR (AA run)
Performance
AA output speedVerified
60 t/s
Median output tokens/s (1k prompt, default provider)
Sources
Where the numbers come from
- Cohere models· checked 2026-06-09
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Command A?
Command A is a Cohere model: Cohere's enterprise-focused Command tier. Strong at RAG, tool use, and multilingual retrieval workflows. 256K context.
What is Command A for?
Pick when your stack already uses Cohere embed + rerank and you want one vendor for retrieval pipelines.
What does Command A cost?
Command A lists $2.5 input and $10 output per 1M tokens. Pricing last checked 2026-06-09.
How large is the Command A context window?
Command A publishes a 256K token context window.
What is the VerdictPal status of Command A?
Command A carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.