Instructions to use patdev/k3-a40-bootstrap 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 patdev/k3-a40-bootstrap 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 patdev/k3-a40-bootstrap:BF16 # Run inference directly in the terminal: llama cli -hf patdev/k3-a40-bootstrap:BF16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf patdev/k3-a40-bootstrap:BF16 # Run inference directly in the terminal: llama cli -hf patdev/k3-a40-bootstrap:BF16
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 patdev/k3-a40-bootstrap:BF16 # Run inference directly in the terminal: ./llama-cli -hf patdev/k3-a40-bootstrap:BF16
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 patdev/k3-a40-bootstrap:BF16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf patdev/k3-a40-bootstrap:BF16
Use Docker
docker model run hf.co/patdev/k3-a40-bootstrap:BF16
- LM Studio
- Jan
- Ollama
How to use patdev/k3-a40-bootstrap with Ollama:
ollama run hf.co/patdev/k3-a40-bootstrap:BF16
- Unsloth Desktop
- Docker Model Runner
How to use patdev/k3-a40-bootstrap with Docker Model Runner:
docker model run hf.co/patdev/k3-a40-bootstrap:BF16
- Lemonade
How to use patdev/k3-a40-bootstrap with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull patdev/k3-a40-bootstrap:BF16
Run and chat with the model
lemonade run user.k3-a40-bootstrap-BF16
List all available models
lemonade list
- Atomic Chat
max-num-seqs 64 : le defaut 256 depasse les blocs Mamba sur 32 Go
Browse files- banc_carte.py +6 -0
banc_carte.py
CHANGED
|
@@ -70,6 +70,12 @@ BASE = ["vllm", "serve", MODEL,
|
|
| 70 |
"--mamba-cache-mode", "align",
|
| 71 |
"--async-scheduling",
|
| 72 |
"--max-num-batched-tokens", "8192",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
"--reasoning-parser", "nemotron_v3",
|
| 74 |
"--tool-call-parser", "qwen3_coder",
|
| 75 |
"--enable-auto-tool-choice"]
|
|
|
|
| 70 |
"--mamba-cache-mode", "align",
|
| 71 |
"--async-scheduling",
|
| 72 |
"--max-num-batched-tokens", "8192",
|
| 73 |
+
# Sur un hybride Mamba, CHAQUE sequence en decodage consomme un bloc de
|
| 74 |
+
# cache Mamba. Le defaut de vLLM (256) depasse ce qu'une carte de 32 Go
|
| 75 |
+
# peut loger -- 117 blocs mesures sur RTX 5090 -- et la capture des
|
| 76 |
+
# graphes CUDA refuse alors de demarrer. 64 suffit largement : le banc
|
| 77 |
+
# ne monte qu'a 8 sessions.
|
| 78 |
+
"--max-num-seqs", os.environ.get("BANC_SEQS", "64"),
|
| 79 |
"--reasoning-parser", "nemotron_v3",
|
| 80 |
"--tool-call-parser", "qwen3_coder",
|
| 81 |
"--enable-auto-tool-choice"]
|