~/model/huggingfaceCohereForAI/c4ai-command-r-plus
Cohere

Command R+

Cohere's 100B-class. Strong RAG and tool-use bench results; non-commercial license.

Parameters
104B
Family
Cohere
Context
128K tokens
FP16 weights
208GB
// where you can run it
command-r-plusLM StudiovLLMMLXoMLX
// hugging face stats (cached daily)
3K downloads · 2K likes · license: cc-by-nc-4.0 · updated 1 year ago

What you need to run this.

$ ./vrambudget --model command-r-plus --by quant
FP16208 GB
M3 Ultra 512512 GB · fitsM2 Ultra 192192 GB · overB200192 GB · overMI300X192 GB · over
Q8_0111 GB
H200141 GB · fits2× H100 NVL188 GB · fitsM2 Ultra 192192 GB · fitsB200192 GB · fits
Q6_K85 GB
M4 Max 128128 GB · fitsM5 Max 128128 GB · fitsDGX Spark128 GB · fitsGaudi 3128 GB · fits
Q5_K_M72 GB
M3 Max 9696 GB · fitsRTX Pro 600096 GB · fitsM4 Max 128128 GB · fitsM5 Max 128128 GB · fits
Q4_K_M59 GB
H100 80GB80 GB · fitsM3 Max 9696 GB · fitsRTX Pro 600096 GB · fitsM4 Max 128128 GB · fits

// budgets shown at ctx 8K, concurrency 1, 15% safety headroom. Tune in the calculator →

Alternatives at this size.

$ grep --params similar catalog.json
$ similargpt-oss 120B117BOpenAI$ similarQwen 2.5 72B72.7BAlibaba$ similarLlama 3.3 70B70.6BMeta$ similarMixtral 8x22B141BMistral

Discussion.

$ gh discussion list

// sign in with github to leave a comment. threads live in the repo's discussions tab.