whatmodel.sh / Command A Reasoning (08-2025)

Command A Reasoning (08-2025)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Command A Reasoning (08-2025) is an open-weights language model from Cohere.

Catalogue identity notes

HF repo sha bd39aafe2c5e71a044d2f1674accfd9b384d3dbb, cc-by-nc-4.0, created 2025-08-12; 111B per the Cohere post's comparison table.

54.8±2.1#270 of 719 LLMsfrom 6 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of Command A Reasoning (08-2025), out of 828 with a score, then more from Cohere.

#modelscorewhereRAMreleased
322Llama 3.1 405B Instruct Meta55±1api $2.40/M +local~160GB2024-07
323Qwen2.5 32B Instruct Alibaba55±1local~21GB2024-09
324Command A Reasoning (08-2025) Cohere55±2local~34GB
325Bielik 11B v3.0 Instruct SpeakLeash55±2local~8.1GB2025-11
326Olmo 3.1 32B Instruct Ai255±2local~21GB2025-12
more from Cohere
76North Small Translate 1.0 Cohere71±6api price unknown2026-08
175Command A+ (05-2026) Cohere64±1api price unknown +local~71GB2026-05
277Command A (03-2025) Cohere57±1api $4.38/M +local~37GB2025-03

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #148 of 300 · 58% of its score · 3 tests

Beats o4-mini, o1 (2024-12-17), o3-mini here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, Jamba 1.5 Large here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
τ²-bench Telecom lab37.0% ✎▁·▅█▆▅▂▃▃▂▃▁ beats 30/89hard · avg 25%▇▇▇▇▇▇▇▇▇▇▇ 27%sharp · at its level · small run
BFCL V4 (December 2025 archive)57.1% ✎▁▁▁▂██▅▅▆▂▂▂ beats 68/77hard · avg 32%▇▇▇▇▇▇▇ 18%at its level
Terminal-Bench Hard lab3.0% ✎▂▅▆▂▂▆▂▂█▅▅▂ beats 4/27very hard · avg 4.2%▇▇▇▇▇ 12%sharp · too hard for it · small run
coding▁▁▁▃▆██▇▆▄▃▁ #183 of 458 * · 8.9% of its score · 1 test

Beats Gemini 2.5 Flash, Qwen3.5 4B, Qwen3 30B-A3B Instruct 2507 here, though they rank higher overall. Trails QwQ-32B-Preview, OpenThinker 32B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
SciCode (level unspecified) lab30.0% ✎▁····▃▄▄▄█▆▂ beats 10/34hard · avg 19%▇▇▇▇ 8.9%small run
math▁▂▃▅▆▅▄█▄▁▁▁ #209 of 398 * · 25% of its score · 1 test

Beats DeepSeek V3 (Dec '24), GPT-4o (2024-08-06), Command A (03-2025) here, though they rank higher overall. Trails Olmo 3 7B Instruct, DeepSeek R1 Distill Llama 8B, Phi-4-mini-reasoning 3.8B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AIME 2025 lab57.0% ✎▁·▁▁▁▂▅▆█▇▅▁ beats 43/133medium · avg 47%▇▇▇▇▇▇▇▇▇▇ 25%sharp · at its level · small run
other tests1 not in a task · 6.7% of its score
testresultvs the others in its rundifficultycountswhy
IFBench prompt-level (loose) lab36.0% ✎▃········██▅ beats 0/8medium · avg 41%▇▇▇ 6.7%at its level · small run

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Command A Reasoning (08-2025) · needs ~67.1 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/CohereLabs_command-a-reasoning-08-2025-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/CohereLabs_command-a-reasoning-08-2025-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Command A Reasoning (08-2025) · needs ~67.1 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/CohereLabs_command-a-reasoning-08-2025-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Command A Reasoning (08-2025) · needs ~30.5 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/CohereLabs_command-a-reasoning-08-2025-GGUF:IQ2_XXS

MLX (Mac)

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Command A Reasoning (08-2025) · needs ~118 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/command-a-reasoning-08-2025-8bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/command-a-reasoning-08-2025-8bit

Weights and documentation: huggingface.co/CohereLabs ↗

$ how the score was worked out · Cohere vendor-reported benchm… 80%, Berkeley Function Calling Lea… 18%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Cohere vendor-reported benchmark scores for its own models (cohere.com blog) · lab-reported5▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 80%Factual vendor-reported score (not licensed content)
Berkeley Function Calling Leaderboard result archive1▇▇▇▇▇▇ 18%Apache-2.0
prior · pull toward the middle▇ 2%
weighted average= 54.8

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json