Text Generation
Safetensors
GGUF
English
mixture-of-experts
Mixture of Experts
deepseek
multi-head-latent-attention
mla
multi-token-prediction
from-scratch
tiny
Instructions to use edededdy/ShallowSeek-mini-base 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 edededdy/ShallowSeek-mini-base 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 edededdy/ShallowSeek-mini-base:F16 # Run inference directly in the terminal: llama cli -hf edededdy/ShallowSeek-mini-base:F16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf edededdy/ShallowSeek-mini-base:F16 # Run inference directly in the terminal: llama cli -hf edededdy/ShallowSeek-mini-base:F16
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf edededdy/ShallowSeek-mini-base:F16 # Run inference directly in the terminal: ./build/bin/llama cli -hf edededdy/ShallowSeek-mini-base:F16
Use Docker
docker model run hf.co/edededdy/ShallowSeek-mini-base:F16
- Llama new
- LM Studio
- Jan
- vLLM
How to use edededdy/ShallowSeek-mini-base with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "edededdy/ShallowSeek-mini-base" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "edededdy/ShallowSeek-mini-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/edededdy/ShallowSeek-mini-base:F16
- Ollama
How to use edededdy/ShallowSeek-mini-base with Ollama:
ollama run hf.co/edededdy/ShallowSeek-mini-base:F16
- Unsloth Desktop
- Docker Model Runner
How to use edededdy/ShallowSeek-mini-base with Docker Model Runner:
docker model run hf.co/edededdy/ShallowSeek-mini-base:F16
- Lemonade
How to use edededdy/ShallowSeek-mini-base with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull edededdy/ShallowSeek-mini-base:F16
Run and chat with the model
lemonade run user.ShallowSeek-mini-base-F16
List all available models
lemonade list
- Atomic Chat
Download config.json from edededdy/ShallowSeek-mini-base: direct link, hf CLI and curl.
- Browser
- Download file 654 Bytes
-
https://huggingface.co/edededdy/ShallowSeek-mini-base/resolve/main/config.json
- Command line
-
hf download hf://edededdy/ShallowSeek-mini-base/config.json
-
curl -L -o config.json https://huggingface.co/edededdy/ShallowSeek-mini-base/resolve/main/config.json
654 Bytes
| { | |
| "vocab_size": 8192, | |
| "max_seq_len": 1024, | |
| "dim": 192, | |
| "n_layers": 8, | |
| "n_dense_layers": 1, | |
| "dense_hidden_dim": 512, | |
| "norm_eps": 1e-06, | |
| "init_std": 0.02, | |
| "n_heads": 6, | |
| "q_lora_rank": 0, | |
| "kv_lora_rank": 48, | |
| "qk_nope_head_dim": 24, | |
| "qk_rope_head_dim": 16, | |
| "v_head_dim": 24, | |
| "rope_theta": 10000.0, | |
| "n_routed_experts": 12, | |
| "n_shared_experts": 1, | |
| "n_activated_experts": 3, | |
| "moe_hidden_dim": 128, | |
| "n_expert_groups": 3, | |
| "n_limited_groups": 2, | |
| "route_scale": 1.0, | |
| "bias_update_speed": 0.001, | |
| "seq_aux_loss_alpha": 0.0001, | |
| "n_mtp_modules": 1, | |
| "mtp_loss_weight": 0.3, | |
| "use_fp8": false, | |
| "fp8_block_size": 128 | |
| } |