Cognitivecomputations Dolphin3.0 R1 Mistral 24B GGUF

quantized_by: bartowski pipeline_tag: text-generation datasets:

• cognitivecomputations/dolphin-r1
• OpenCoder-LLM/opc-sft-stage1
• OpenCoder-LLM/opc-sft-stage2
• microsoft/orca-agentinstruct-1M-v1
• microsoft/orca-math-word-problems-200k
• NousResearch/hermes-function-calling-v1
• AI-MO/NuminaMath-CoT
• AI-MO/NuminaMath-TIR
• allenai/tulu-3-sft-mixture
• cognitivecomputations/dolphin-coder
• HuggingFaceTB/smoltalk
• cognitivecomputations/samantha-data
• m-a-p/CodeFeedback-Filtered-Instruction
• m-a-p/Code-Feedback base_model: cognitivecomputations/Dolphin3.0-R1-Mistral-24B language:
• en

Llamacpp imatrix Quantizations of Dolphin3.0-R1-Mistral-24B by cognitivecomputations

Using llama.cpp release b4585 for quantization.

Original model: https://huggingface.co/cognitivecomputations/Dolphin3.0-R1-Mistral-24B

All quants made using imatrix option with dataset from here

Run them in LM Studio

Run them directly with llama.cpp, or any other llama.cpp based project

Prompt format

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Recommended reasoning system prompt

For reasoning, it's recommended you use the following system prompt:

You are Dolphin, an AI assistant that helps humanity, trained by Eric Hartford to specialize in reasoning and first-principles analysis.

When responding, always format your replies using <think>{reasoning}</think>{answer}. Use at least 6 reasoning steps and perform a root cause analysis before answering. However, if the answer is very easy and requires little thought, you may leave the <think></think> block empty.

Your responses should be detailed, structured with rich Markdown formatting, and engaging with emojis. Be extensive in your explanations, just as the greatest scientific minds would be. Always reason through the problem first, unless it's trivial, in which case you may answer directly.

Download a file (not the whole branch) from below:

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

Downloading using huggingface-cli

pip install -U "huggingface_hub[cli]"

huggingface-cli download bartowski/cognitivecomputations_Dolphin3.0-R1-Mistral-24B-GGUF --include "cognitivecomputations_Dolphin3.0-R1-Mistral-24B-Q4_K_M.gguf" --local-dir ./

huggingface-cli download bartowski/cognitivecomputations_Dolphin3.0-R1-Mistral-24B-GGUF --include "cognitivecomputations_Dolphin3.0-R1-Mistral-24B-Q8_0/*" --local-dir ./

ARM/AVX information

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Which file should I choose?

Credits

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.

Thank you ZeroWw for the inspiration to experiment with embed/output.

Thank you to LM Studio for sponsoring my work.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski