Cohere
Cohere
Multimodal model · Canada ·May 2026

Command A+

Multimodal Open Weights
218B
Parameters
128K
Context
2
Benchmarks

About Command A+

Command A+ is an AI model developed by Cohere, in the multimodal category, released in 2026, made available as an open-weights model with 218B params. Command A+ is Cohere's fastest and most powerful model, a 218B-total / 25B-active sparse Mixture-of-Experts released und.

On this page you'll find Command A+'s full specifications, including a context window of 128K tokens, 2 benchmark results. Review provider pricing and benchmark scores below, or compare Command A+ head-to-head with other multimodal models.

→ More models by Cohere → All multimodal models → AI benchmarks leaderboard

Run Command A+ locally — what to buy

Share this build ↗

Sized to what Command A+ actually needs (~128k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Best
Apple M2 Ultra 192GB
Apple · ~$6,600
~5.4 tok/s
Q4 · fits full context · up to 94k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Apple
5.4 tok/s
1 option →
Datacenter GPU
33.4–57 tok/s
4 options →

Community

Would you run Command A+ again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Command A+ benchmark scores▶
Compare on the benchmark table →
Benchmark Category Score Variant
MMMU multimodal 75.1%
MathVista multimodal 80.6%

More models like Command A+

AI
NeuroChess
· 72.3K
 Cohere
North Mini Code 1.0
Cohere · 30B
 Cohere
Cohere Transcribe
Cohere · 2B
 Cohere
Command A Translate
Cohere
 Cohere
Command A Reasoning
Cohere
 Cohere
Cohere Command A Reasoning
Cohere · 111B

Frequently asked questions

What is Command A+? ▶

Command A+ is Cohere's fastest and most powerful model, a 218B-total / 25B-active sparse Mixture-of-Experts released und It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Command A+? ▶

Command A+ was developed by Cohere and released in 2026.

Is Command A+ open source or proprietary? ▶

Command A+ ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

What is Command A+'s context window? ▶

Command A+ supports a context window of 128K tokens, and has 218B params.

How much does Command A+ cost? ▶

Pricing for Command A+ depends on the provider. See the providers table on this page for the latest API rates.

How does Command A+ perform on benchmarks? ▶

Command A+ is benchmarked across 2 evaluations on GenAIList, including MMMU (75.1%). See the full benchmark table below and compare it with other models.

Command A+ : Cohere's Best Model Ever Is Kind of Disappointing
This NEW Cohere Command A+ is a GAME CHANGER!

Know a great video about Command A+?