Command R Plus
104B
490 Pulls Updated 5 weeks ago
Updated 5 weeks ago
5 weeks ago
ff3470d405e3 · 51GB
model
51GB
template
<BOS_TOKEN>{{ if .System }}<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>{{ .System }}<|END_OF_TURN_TOKEN|>{{ end }}<|START_OF_TURN_TOKEN|><|USER_TOKEN|>{{ .Prompt }}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>{{ .Response }}
235B
params
{"stop":["<|END_OF_TURN_TOKEN|>"]}
45B
license
CC-BY-NC 4.0 License with Acceptable Use Addendum https://cohere.com/c4ai-cc-by-nc-license
90B
Readme
Command R Plus
Model: https://huggingface.co/CohereForAI/c4ai-command-r-plus
I converted to FP16, then quantised (and made use of dranger003’s iMatrix which is trained on 100K~ of https://huggingface.co/datasets/wikitext).
Note Command R Plus currently requires building Ollama with the llama.cpp branch from this PR: https://github.com/ggerganov/llama.cpp/pull/6491