Instructions to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Use Docker
docker model run hf.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "JetBrains/Qwen3.8-3.6-27B-blend-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "JetBrains/Qwen3.8-3.6-27B-blend-GGUF", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
- Ollama
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with Ollama:
ollama run hf.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
- Unsloth Desktop
- Pi
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with Docker Model Runner:
docker model run hf.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
- Lemonade
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.Qwen3.8-3.6-27B-blend-GGUF-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use JetBrains/Qwen3.8-3.6-27B-blend-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "JetBrains/Qwen3.8-3.6-27B-blend-GGUF:Q4_K_M" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Download Q3_K_XL.tensor-types.txt from JetBrains/Qwen3.8-3.6-27B-blend-GGUF: direct link, hf CLI and curl.
- Browser
- Download file 17 kB
-
https://huggingface.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF/resolve/main/Q3_K_XL.tensor-types.txt
- Command line
-
hf download hf://JetBrains/Qwen3.8-3.6-27B-blend-GGUF/Q3_K_XL.tensor-types.txt
-
curl -L -o Q3_K_XL.tensor-types.txt https://huggingface.co/JetBrains/Qwen3.8-3.6-27B-blend-GGUF/resolve/main/Q3_K_XL.tensor-types.txt
17 kB
| ^output\.weight$=q5_k | |
| ^token_embd\.weight$=q3_k | |
| ^blk\.0\.attn_gate\.weight$=iq4_xs | |
| ^blk\.0\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.0\.ffn_down\.weight$=iq3_xxs | |
| ^blk\.0\.ffn_gate\.weight$=iq2_xs | |
| ^blk\.0\.ffn_up\.weight$=iq2_s | |
| ^blk\.0\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.0\.ssm_beta\.weight$=q8_0 | |
| ^blk\.0\.ssm_out\.weight$=q5_k | |
| ^blk\.1\.attn_gate\.weight$=q4_k | |
| ^blk\.1\.attn_qkv\.weight$=q4_k | |
| ^blk\.1\.ffn_down\.weight$=iq3_s | |
| ^blk\.1\.ffn_gate\.weight$=iq2_s | |
| ^blk\.1\.ffn_up\.weight$=q3_k | |
| ^blk\.1\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.1\.ssm_beta\.weight$=q8_0 | |
| ^blk\.1\.ssm_out\.weight$=q4_k | |
| ^blk\.2\.attn_gate\.weight$=q4_k | |
| ^blk\.2\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.2\.ffn_down\.weight$=iq4_xs | |
| ^blk\.2\.ffn_gate\.weight$=iq3_s | |
| ^blk\.2\.ffn_up\.weight$=iq3_s | |
| ^blk\.2\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.2\.ssm_beta\.weight$=q8_0 | |
| ^blk\.2\.ssm_out\.weight$=iq4_xs | |
| ^blk\.3\.attn_k\.weight$=q4_k | |
| ^blk\.3\.attn_output\.weight$=q4_k | |
| ^blk\.3\.attn_q\.weight$=iq4_nl | |
| ^blk\.3\.attn_v\.weight$=q5_k | |
| ^blk\.3\.ffn_down\.weight$=iq4_xs | |
| ^blk\.3\.ffn_gate\.weight$=q3_k | |
| ^blk\.3\.ffn_up\.weight$=iq3_s | |
| ^blk\.4\.attn_gate\.weight$=q4_k | |
| ^blk\.4\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.4\.ffn_down\.weight$=q4_k | |
| ^blk\.4\.ffn_gate\.weight$=iq3_s | |
| ^blk\.4\.ffn_up\.weight$=iq3_s | |
| ^blk\.4\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.4\.ssm_beta\.weight$=q8_0 | |
| ^blk\.4\.ssm_out\.weight$=q4_k | |
| ^blk\.5\.attn_gate\.weight$=iq4_xs | |
| ^blk\.5\.attn_qkv\.weight$=iq3_s | |
| ^blk\.5\.ffn_down\.weight$=iq3_s | |
| ^blk\.5\.ffn_gate\.weight$=iq3_s | |
| ^blk\.5\.ffn_up\.weight$=iq3_s | |
| ^blk\.5\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.5\.ssm_beta\.weight$=q8_0 | |
| ^blk\.5\.ssm_out\.weight$=q4_k | |
| ^blk\.6\.attn_gate\.weight$=q4_k | |
| ^blk\.6\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.6\.ffn_down\.weight$=iq4_xs | |
| ^blk\.6\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.6\.ffn_up\.weight$=iq3_s | |
| ^blk\.6\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.6\.ssm_beta\.weight$=q8_0 | |
| ^blk\.6\.ssm_out\.weight$=iq4_xs | |
| ^blk\.7\.attn_k\.weight$=q5_k | |
| ^blk\.7\.attn_output\.weight$=iq4_xs | |
| ^blk\.7\.attn_q\.weight$=q2_k | |
| ^blk\.7\.attn_v\.weight$=q5_k | |
| ^blk\.7\.ffn_down\.weight$=q3_k | |
| ^blk\.7\.ffn_gate\.weight$=q3_k | |
| ^blk\.7\.ffn_up\.weight$=iq3_s | |
| ^blk\.8\.attn_gate\.weight$=iq4_xs | |
| ^blk\.8\.attn_qkv\.weight$=iq3_s | |
| ^blk\.8\.ffn_down\.weight$=iq3_s | |
| ^blk\.8\.ffn_gate\.weight$=iq2_s | |
| ^blk\.8\.ffn_up\.weight$=iq3_s | |
| ^blk\.8\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.8\.ssm_beta\.weight$=q8_0 | |
| ^blk\.8\.ssm_out\.weight$=q4_k | |
| ^blk\.9\.attn_gate\.weight$=iq3_s | |
| ^blk\.9\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.9\.ffn_down\.weight$=iq3_s | |
| ^blk\.9\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.9\.ffn_up\.weight$=iq3_s | |
| ^blk\.9\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.9\.ssm_beta\.weight$=q8_0 | |
| ^blk\.9\.ssm_out\.weight$=iq4_xs | |
| ^blk\.10\.attn_gate\.weight$=iq3_xxs | |
| ^blk\.10\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.10\.ffn_down\.weight$=iq3_xxs | |
| ^blk\.10\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.10\.ffn_up\.weight$=iq2_s | |
| ^blk\.10\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.10\.ssm_beta\.weight$=q8_0 | |
| ^blk\.10\.ssm_out\.weight$=q3_k | |
| ^blk\.11\.attn_k\.weight$=iq4_nl | |
| ^blk\.11\.attn_output\.weight$=iq3_xxs | |
| ^blk\.11\.attn_q\.weight$=q2_k | |
| ^blk\.11\.attn_v\.weight$=q4_k | |
| ^blk\.11\.ffn_down\.weight$=iq3_xxs | |
| ^blk\.11\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.11\.ffn_up\.weight$=iq2_s | |
| ^blk\.12\.attn_gate\.weight$=iq3_xxs | |
| ^blk\.12\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.12\.ffn_down\.weight$=iq2_s | |
| ^blk\.12\.ffn_gate\.weight$=iq2_xs | |
| ^blk\.12\.ffn_up\.weight$=iq2_s | |
| ^blk\.12\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.12\.ssm_beta\.weight$=q8_0 | |
| ^blk\.12\.ssm_out\.weight$=q3_k | |
| ^blk\.13\.attn_gate\.weight$=iq2_s | |
| ^blk\.13\.attn_qkv\.weight$=iq3_s | |
| ^blk\.13\.ffn_down\.weight$=iq2_s | |
| ^blk\.13\.ffn_gate\.weight$=iq2_xs | |
| ^blk\.13\.ffn_up\.weight$=iq3_xxs | |
| ^blk\.13\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.13\.ssm_beta\.weight$=q8_0 | |
| ^blk\.13\.ssm_out\.weight$=iq3_s | |
| ^blk\.14\.attn_gate\.weight$=iq3_s | |
| ^blk\.14\.attn_qkv\.weight$=iq3_s | |
| ^blk\.14\.ffn_down\.weight$=iq2_xs | |
| ^blk\.14\.ffn_gate\.weight$=iq2_xxs | |
| ^blk\.14\.ffn_up\.weight$=iq2_s | |
| ^blk\.14\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.14\.ssm_beta\.weight$=q8_0 | |
| ^blk\.14\.ssm_out\.weight$=iq3_xxs | |
| ^blk\.15\.attn_k\.weight$=q4_k | |
| ^blk\.15\.attn_output\.weight$=q3_k | |
| ^blk\.15\.attn_q\.weight$=q2_k | |
| ^blk\.15\.attn_v\.weight$=q5_k | |
| ^blk\.15\.ffn_down\.weight$=iq2_s | |
| ^blk\.15\.ffn_gate\.weight$=iq2_xxs | |
| ^blk\.15\.ffn_up\.weight$=iq2_s | |
| ^blk\.16\.attn_gate\.weight$=q4_k | |
| ^blk\.16\.attn_qkv\.weight$=iq3_s | |
| ^blk\.16\.ffn_down\.weight$=iq2_s | |
| ^blk\.16\.ffn_gate\.weight$=iq2_s | |
| ^blk\.16\.ffn_up\.weight$=iq3_xxs | |
| ^blk\.16\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.16\.ssm_beta\.weight$=q8_0 | |
| ^blk\.16\.ssm_out\.weight$=iq4_xs | |
| ^blk\.17\.attn_gate\.weight$=iq4_xs | |
| ^blk\.17\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.17\.ffn_down\.weight$=iq3_xxs | |
| ^blk\.17\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.17\.ffn_up\.weight$=iq2_s | |
| ^blk\.17\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.17\.ssm_beta\.weight$=q8_0 | |
| ^blk\.17\.ssm_out\.weight$=q3_k | |
| ^blk\.18\.attn_gate\.weight$=iq3_s | |
| ^blk\.18\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.18\.ffn_down\.weight$=iq3_s | |
| ^blk\.18\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.18\.ffn_up\.weight$=iq3_xxs | |
| ^blk\.18\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.18\.ssm_beta\.weight$=q8_0 | |
| ^blk\.18\.ssm_out\.weight$=iq4_xs | |
| ^blk\.19\.attn_k\.weight$=q4_k | |
| ^blk\.19\.attn_output\.weight$=iq3_s | |
| ^blk\.19\.attn_q\.weight$=iq3_xxs | |
| ^blk\.19\.attn_v\.weight$=q5_k | |
| ^blk\.19\.ffn_down\.weight$=iq3_s | |
| ^blk\.19\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.19\.ffn_up\.weight$=iq3_xxs | |
| ^blk\.20\.attn_gate\.weight$=iq4_xs | |
| ^blk\.20\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.20\.ffn_down\.weight$=iq4_xs | |
| ^blk\.20\.ffn_gate\.weight$=iq3_s | |
| ^blk\.20\.ffn_up\.weight$=iq3_s | |
| ^blk\.20\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.20\.ssm_beta\.weight$=q8_0 | |
| ^blk\.20\.ssm_out\.weight$=q4_k | |
| ^blk\.21\.attn_gate\.weight$=q4_k | |
| ^blk\.21\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.21\.ffn_down\.weight$=iq4_xs | |
| ^blk\.21\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.21\.ffn_up\.weight$=q3_k | |
| ^blk\.21\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.21\.ssm_beta\.weight$=q8_0 | |
| ^blk\.21\.ssm_out\.weight$=iq4_xs | |
| ^blk\.22\.attn_gate\.weight$=iq4_xs | |
| ^blk\.22\.attn_qkv\.weight$=iq3_s | |
| ^blk\.22\.ffn_down\.weight$=iq4_xs | |
| ^blk\.22\.ffn_gate\.weight$=iq3_s | |
| ^blk\.22\.ffn_up\.weight$=iq3_s | |
| ^blk\.22\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.22\.ssm_beta\.weight$=q8_0 | |
| ^blk\.22\.ssm_out\.weight$=iq4_xs | |
| ^blk\.23\.attn_k\.weight$=q5_k | |
| ^blk\.23\.attn_output\.weight$=q4_k | |
| ^blk\.23\.attn_q\.weight$=iq3_s | |
| ^blk\.23\.attn_v\.weight$=q5_k | |
| ^blk\.23\.ffn_down\.weight$=iq4_xs | |
| ^blk\.23\.ffn_gate\.weight$=iq3_s | |
| ^blk\.23\.ffn_up\.weight$=iq4_xs | |
| ^blk\.24\.attn_gate\.weight$=iq4_xs | |
| ^blk\.24\.attn_qkv\.weight$=iq3_s | |
| ^blk\.24\.ffn_down\.weight$=iq4_xs | |
| ^blk\.24\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.24\.ffn_up\.weight$=q4_k | |
| ^blk\.24\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.24\.ssm_beta\.weight$=q8_0 | |
| ^blk\.24\.ssm_out\.weight$=iq4_xs | |
| ^blk\.25\.attn_gate\.weight$=iq4_xs | |
| ^blk\.25\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.25\.ffn_down\.weight$=iq4_xs | |
| ^blk\.25\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.25\.ffn_up\.weight$=iq4_xs | |
| ^blk\.25\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.25\.ssm_beta\.weight$=q8_0 | |
| ^blk\.25\.ssm_out\.weight$=iq4_xs | |
| ^blk\.26\.attn_gate\.weight$=iq4_xs | |
| ^blk\.26\.attn_qkv\.weight$=iq3_s | |
| ^blk\.26\.ffn_down\.weight$=iq4_xs | |
| ^blk\.26\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.26\.ffn_up\.weight$=iq4_xs | |
| ^blk\.26\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.26\.ssm_beta\.weight$=q8_0 | |
| ^blk\.26\.ssm_out\.weight$=iq4_xs | |
| ^blk\.27\.attn_k\.weight$=q5_k | |
| ^blk\.27\.attn_output\.weight$=iq4_xs | |
| ^blk\.27\.attn_q\.weight$=iq4_xs | |
| ^blk\.27\.attn_v\.weight$=q5_k | |
| ^blk\.27\.ffn_down\.weight$=iq4_xs | |
| ^blk\.27\.ffn_gate\.weight$=iq3_s | |
| ^blk\.27\.ffn_up\.weight$=iq3_s | |
| ^blk\.28\.attn_gate\.weight$=q4_k | |
| ^blk\.28\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.28\.ffn_down\.weight$=q3_k | |
| ^blk\.28\.ffn_gate\.weight$=iq3_s | |
| ^blk\.28\.ffn_up\.weight$=iq4_xs | |
| ^blk\.28\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.28\.ssm_beta\.weight$=q8_0 | |
| ^blk\.28\.ssm_out\.weight$=iq4_xs | |
| ^blk\.29\.attn_gate\.weight$=q4_k | |
| ^blk\.29\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.29\.ffn_down\.weight$=iq3_s | |
| ^blk\.29\.ffn_gate\.weight$=iq3_s | |
| ^blk\.29\.ffn_up\.weight$=iq3_s | |
| ^blk\.29\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.29\.ssm_beta\.weight$=q8_0 | |
| ^blk\.29\.ssm_out\.weight$=q4_k | |
| ^blk\.30\.attn_gate\.weight$=iq3_s | |
| ^blk\.30\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.30\.ffn_down\.weight$=iq3_s | |
| ^blk\.30\.ffn_gate\.weight$=iq3_s | |
| ^blk\.30\.ffn_up\.weight$=iq3_s | |
| ^blk\.30\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.30\.ssm_beta\.weight$=q8_0 | |
| ^blk\.30\.ssm_out\.weight$=iq4_xs | |
| ^blk\.31\.attn_k\.weight$=q4_k | |
| ^blk\.31\.attn_output\.weight$=iq4_xs | |
| ^blk\.31\.attn_q\.weight$=iq4_xs | |
| ^blk\.31\.attn_v\.weight$=q5_k | |
| ^blk\.31\.ffn_down\.weight$=iq3_s | |
| ^blk\.31\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.31\.ffn_up\.weight$=iq3_s | |
| ^blk\.32\.attn_gate\.weight$=iq3_xxs | |
| ^blk\.32\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.32\.ffn_down\.weight$=iq3_s | |
| ^blk\.32\.ffn_gate\.weight$=iq3_s | |
| ^blk\.32\.ffn_up\.weight$=iq3_s | |
| ^blk\.32\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.32\.ssm_beta\.weight$=q8_0 | |
| ^blk\.32\.ssm_out\.weight$=iq4_xs | |
| ^blk\.33\.attn_gate\.weight$=iq4_xs | |
| ^blk\.33\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.33\.ffn_down\.weight$=iq3_s | |
| ^blk\.33\.ffn_gate\.weight$=iq3_s | |
| ^blk\.33\.ffn_up\.weight$=iq3_s | |
| ^blk\.33\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.33\.ssm_beta\.weight$=q8_0 | |
| ^blk\.33\.ssm_out\.weight$=iq4_xs | |
| ^blk\.34\.attn_gate\.weight$=iq4_xs | |
| ^blk\.34\.attn_qkv\.weight$=iq3_s | |
| ^blk\.34\.ffn_down\.weight$=iq3_s | |
| ^blk\.34\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.34\.ffn_up\.weight$=iq3_s | |
| ^blk\.34\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.34\.ssm_beta\.weight$=q8_0 | |
| ^blk\.34\.ssm_out\.weight$=iq4_xs | |
| ^blk\.35\.attn_k\.weight$=q4_k | |
| ^blk\.35\.attn_output\.weight$=q4_k | |
| ^blk\.35\.attn_q\.weight$=iq3_s | |
| ^blk\.35\.attn_v\.weight$=q5_k | |
| ^blk\.35\.ffn_down\.weight$=iq4_xs | |
| ^blk\.35\.ffn_gate\.weight$=iq3_s | |
| ^blk\.35\.ffn_up\.weight$=iq3_s | |
| ^blk\.36\.attn_gate\.weight$=iq4_xs | |
| ^blk\.36\.attn_qkv\.weight$=iq3_s | |
| ^blk\.36\.ffn_down\.weight$=iq4_xs | |
| ^blk\.36\.ffn_gate\.weight$=iq3_s | |
| ^blk\.36\.ffn_up\.weight$=iq4_xs | |
| ^blk\.36\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.36\.ssm_beta\.weight$=q8_0 | |
| ^blk\.36\.ssm_out\.weight$=iq4_xs | |
| ^blk\.37\.attn_gate\.weight$=iq4_xs | |
| ^blk\.37\.attn_qkv\.weight$=iq3_xxs | |
| ^blk\.37\.ffn_down\.weight$=iq4_xs | |
| ^blk\.37\.ffn_gate\.weight$=iq3_s | |
| ^blk\.37\.ffn_up\.weight$=iq3_s | |
| ^blk\.37\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.37\.ssm_beta\.weight$=q8_0 | |
| ^blk\.37\.ssm_out\.weight$=iq4_xs | |
| ^blk\.38\.attn_gate\.weight$=iq3_s | |
| ^blk\.38\.attn_qkv\.weight$=iq3_s | |
| ^blk\.38\.ffn_down\.weight$=iq4_xs | |
| ^blk\.38\.ffn_gate\.weight$=iq3_s | |
| ^blk\.38\.ffn_up\.weight$=iq3_s | |
| ^blk\.38\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.38\.ssm_beta\.weight$=q8_0 | |
| ^blk\.38\.ssm_out\.weight$=iq3_s | |
| ^blk\.39\.attn_k\.weight$=q4_k | |
| ^blk\.39\.attn_output\.weight$=iq4_xs | |
| ^blk\.39\.attn_q\.weight$=iq3_s | |
| ^blk\.39\.attn_v\.weight$=q5_k | |
| ^blk\.39\.ffn_down\.weight$=iq4_xs | |
| ^blk\.39\.ffn_gate\.weight$=iq3_s | |
| ^blk\.39\.ffn_up\.weight$=iq4_xs | |
| ^blk\.40\.attn_gate\.weight$=iq4_xs | |
| ^blk\.40\.attn_qkv\.weight$=iq3_s | |
| ^blk\.40\.ffn_down\.weight$=iq4_xs | |
| ^blk\.40\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.40\.ffn_up\.weight$=iq4_xs | |
| ^blk\.40\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.40\.ssm_beta\.weight$=q8_0 | |
| ^blk\.40\.ssm_out\.weight$=iq4_xs | |
| ^blk\.41\.attn_gate\.weight$=iq4_xs | |
| ^blk\.41\.attn_qkv\.weight$=iq3_s | |
| ^blk\.41\.ffn_down\.weight$=iq4_xs | |
| ^blk\.41\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.41\.ffn_up\.weight$=iq4_xs | |
| ^blk\.41\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.41\.ssm_beta\.weight$=q8_0 | |
| ^blk\.41\.ssm_out\.weight$=iq4_xs | |
| ^blk\.42\.attn_gate\.weight$=iq4_xs | |
| ^blk\.42\.attn_qkv\.weight$=iq3_s | |
| ^blk\.42\.ffn_down\.weight$=iq4_xs | |
| ^blk\.42\.ffn_gate\.weight$=iq3_s | |
| ^blk\.42\.ffn_up\.weight$=iq4_xs | |
| ^blk\.42\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.42\.ssm_beta\.weight$=q8_0 | |
| ^blk\.42\.ssm_out\.weight$=iq4_xs | |
| ^blk\.43\.attn_k\.weight$=q4_k | |
| ^blk\.43\.attn_output\.weight$=q4_k | |
| ^blk\.43\.attn_q\.weight$=iq3_s | |
| ^blk\.43\.attn_v\.weight$=q5_k | |
| ^blk\.43\.ffn_down\.weight$=iq4_xs | |
| ^blk\.43\.ffn_gate\.weight$=iq3_s | |
| ^blk\.43\.ffn_up\.weight$=iq3_s | |
| ^blk\.44\.attn_gate\.weight$=iq3_s | |
| ^blk\.44\.attn_qkv\.weight$=iq3_s | |
| ^blk\.44\.ffn_down\.weight$=iq3_s | |
| ^blk\.44\.ffn_gate\.weight$=iq3_s | |
| ^blk\.44\.ffn_up\.weight$=iq3_s | |
| ^blk\.44\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.44\.ssm_beta\.weight$=q8_0 | |
| ^blk\.44\.ssm_out\.weight$=iq4_xs | |
| ^blk\.45\.attn_gate\.weight$=iq3_xxs | |
| ^blk\.45\.attn_qkv\.weight$=iq3_s | |
| ^blk\.45\.ffn_down\.weight$=iq3_s | |
| ^blk\.45\.ffn_gate\.weight$=iq3_s | |
| ^blk\.45\.ffn_up\.weight$=iq3_s | |
| ^blk\.45\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.45\.ssm_beta\.weight$=q8_0 | |
| ^blk\.45\.ssm_out\.weight$=iq3_s | |
| ^blk\.46\.attn_gate\.weight$=iq3_s | |
| ^blk\.46\.attn_qkv\.weight$=iq3_s | |
| ^blk\.46\.ffn_down\.weight$=iq3_s | |
| ^blk\.46\.ffn_gate\.weight$=iq3_s | |
| ^blk\.46\.ffn_up\.weight$=iq3_s | |
| ^blk\.46\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.46\.ssm_beta\.weight$=q8_0 | |
| ^blk\.46\.ssm_out\.weight$=iq4_xs | |
| ^blk\.47\.attn_k\.weight$=q4_k | |
| ^blk\.47\.attn_output\.weight$=iq4_xs | |
| ^blk\.47\.attn_q\.weight$=iq3_xxs | |
| ^blk\.47\.attn_v\.weight$=q5_k | |
| ^blk\.47\.ffn_down\.weight$=q3_k | |
| ^blk\.47\.ffn_gate\.weight$=iq3_s | |
| ^blk\.47\.ffn_up\.weight$=iq3_s | |
| ^blk\.48\.attn_gate\.weight$=iq3_s | |
| ^blk\.48\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.48\.ffn_down\.weight$=iq3_s | |
| ^blk\.48\.ffn_gate\.weight$=iq3_s | |
| ^blk\.48\.ffn_up\.weight$=iq3_s | |
| ^blk\.48\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.48\.ssm_beta\.weight$=q8_0 | |
| ^blk\.48\.ssm_out\.weight$=iq4_xs | |
| ^blk\.49\.attn_gate\.weight$=iq3_xxs | |
| ^blk\.49\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.49\.ffn_down\.weight$=iq3_s | |
| ^blk\.49\.ffn_gate\.weight$=iq3_xxs | |
| ^blk\.49\.ffn_up\.weight$=iq3_s | |
| ^blk\.49\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.49\.ssm_beta\.weight$=q8_0 | |
| ^blk\.49\.ssm_out\.weight$=iq4_xs | |
| ^blk\.50\.attn_gate\.weight$=iq4_xs | |
| ^blk\.50\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.50\.ffn_down\.weight$=iq4_xs | |
| ^blk\.50\.ffn_gate\.weight$=iq3_s | |
| ^blk\.50\.ffn_up\.weight$=iq4_xs | |
| ^blk\.50\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.50\.ssm_beta\.weight$=q8_0 | |
| ^blk\.50\.ssm_out\.weight$=iq4_xs | |
| ^blk\.51\.attn_k\.weight$=q5_k | |
| ^blk\.51\.attn_output\.weight$=iq4_xs | |
| ^blk\.51\.attn_q\.weight$=iq3_s | |
| ^blk\.51\.attn_v\.weight$=q5_k | |
| ^blk\.51\.ffn_down\.weight$=iq4_xs | |
| ^blk\.51\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.51\.ffn_up\.weight$=iq4_xs | |
| ^blk\.52\.attn_gate\.weight$=iq4_xs | |
| ^blk\.52\.attn_qkv\.weight$=q4_k | |
| ^blk\.52\.ffn_down\.weight$=iq4_xs | |
| ^blk\.52\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.52\.ffn_up\.weight$=iq4_xs | |
| ^blk\.52\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.52\.ssm_beta\.weight$=q8_0 | |
| ^blk\.52\.ssm_out\.weight$=iq4_xs | |
| ^blk\.53\.attn_gate\.weight$=iq4_xs | |
| ^blk\.53\.attn_qkv\.weight$=iq3_s | |
| ^blk\.53\.ffn_down\.weight$=iq4_xs | |
| ^blk\.53\.ffn_gate\.weight$=iq3_s | |
| ^blk\.53\.ffn_up\.weight$=iq4_xs | |
| ^blk\.53\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.53\.ssm_beta\.weight$=q8_0 | |
| ^blk\.53\.ssm_out\.weight$=q4_k | |
| ^blk\.54\.attn_gate\.weight$=iq4_xs | |
| ^blk\.54\.attn_qkv\.weight$=iq3_s | |
| ^blk\.54\.ffn_down\.weight$=iq4_xs | |
| ^blk\.54\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.54\.ffn_up\.weight$=iq4_xs | |
| ^blk\.54\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.54\.ssm_beta\.weight$=q8_0 | |
| ^blk\.54\.ssm_out\.weight$=iq4_xs | |
| ^blk\.55\.attn_k\.weight$=q4_k | |
| ^blk\.55\.attn_output\.weight$=iq4_xs | |
| ^blk\.55\.attn_q\.weight$=iq3_s | |
| ^blk\.55\.attn_v\.weight$=q5_k | |
| ^blk\.55\.ffn_down\.weight$=iq4_xs | |
| ^blk\.55\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.55\.ffn_up\.weight$=iq4_xs | |
| ^blk\.56\.attn_gate\.weight$=iq4_xs | |
| ^blk\.56\.attn_qkv\.weight$=iq3_s | |
| ^blk\.56\.ffn_down\.weight$=q5_k | |
| ^blk\.56\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.56\.ffn_up\.weight$=iq4_xs | |
| ^blk\.56\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.56\.ssm_beta\.weight$=q8_0 | |
| ^blk\.56\.ssm_out\.weight$=iq4_xs | |
| ^blk\.57\.attn_gate\.weight$=iq4_xs | |
| ^blk\.57\.attn_qkv\.weight$=iq3_s | |
| ^blk\.57\.ffn_down\.weight$=iq4_xs | |
| ^blk\.57\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.57\.ffn_up\.weight$=iq4_xs | |
| ^blk\.57\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.57\.ssm_beta\.weight$=q8_0 | |
| ^blk\.57\.ssm_out\.weight$=iq4_xs | |
| ^blk\.58\.attn_gate\.weight$=iq4_xs | |
| ^blk\.58\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.58\.ffn_down\.weight$=iq4_xs | |
| ^blk\.58\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.58\.ffn_up\.weight$=iq4_xs | |
| ^blk\.58\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.58\.ssm_beta\.weight$=q8_0 | |
| ^blk\.58\.ssm_out\.weight$=iq4_xs | |
| ^blk\.59\.attn_k\.weight$=q5_k | |
| ^blk\.59\.attn_output\.weight$=iq4_xs | |
| ^blk\.59\.attn_q\.weight$=iq3_s | |
| ^blk\.59\.attn_v\.weight$=q5_k | |
| ^blk\.59\.ffn_down\.weight$=iq4_xs | |
| ^blk\.59\.ffn_gate\.weight$=q4_k | |
| ^blk\.59\.ffn_up\.weight$=iq4_xs | |
| ^blk\.60\.attn_gate\.weight$=iq4_xs | |
| ^blk\.60\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.60\.ffn_down\.weight$=iq4_xs | |
| ^blk\.60\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.60\.ffn_up\.weight$=iq4_xs | |
| ^blk\.60\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.60\.ssm_beta\.weight$=q8_0 | |
| ^blk\.60\.ssm_out\.weight$=iq4_xs | |
| ^blk\.61\.attn_gate\.weight$=iq4_xs | |
| ^blk\.61\.attn_qkv\.weight$=iq3_s | |
| ^blk\.61\.ffn_down\.weight$=iq4_xs | |
| ^blk\.61\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.61\.ffn_up\.weight$=iq4_xs | |
| ^blk\.61\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.61\.ssm_beta\.weight$=q8_0 | |
| ^blk\.61\.ssm_out\.weight$=iq4_xs | |
| ^blk\.62\.attn_gate\.weight$=iq4_xs | |
| ^blk\.62\.attn_qkv\.weight$=iq4_xs | |
| ^blk\.62\.ffn_down\.weight$=iq4_xs | |
| ^blk\.62\.ffn_gate\.weight$=iq4_xs | |
| ^blk\.62\.ffn_up\.weight$=iq4_xs | |
| ^blk\.62\.ssm_alpha\.weight$=q8_0 | |
| ^blk\.62\.ssm_beta\.weight$=q8_0 | |
| ^blk\.62\.ssm_out\.weight$=q4_k | |
| ^blk\.63\.attn_k\.weight$=q5_k | |
| ^blk\.63\.attn_output\.weight$=q5_k | |
| ^blk\.63\.attn_q\.weight$=iq4_xs | |
| ^blk\.63\.attn_v\.weight$=q6_k | |
| ^blk\.63\.ffn_down\.weight$=q4_k | |
| ^blk\.63\.ffn_gate\.weight$=q5_k | |
| ^blk\.63\.ffn_up\.weight$=q5_k | |
| ^blk\.64\.attn_k\.weight$=q8_0 | |
| ^blk\.64\.attn_output\.weight$=q6_k | |
| ^blk\.64\.attn_q\.weight$=q6_k | |
| ^blk\.64\.attn_v\.weight$=q8_0 | |
| ^blk\.64\.ffn_down\.weight$=q6_k | |
| ^blk\.64\.ffn_gate\.weight$=q6_k | |
| ^blk\.64\.ffn_up\.weight$=q6_k | |
| ^blk\.64\.nextn\.eh_proj\.weight$=q6_k | |