Coding
How to deploy deepseek-coder-6.7b-instruct on a GPU cloud
A 6.7B model tuned for code generation. Full specs, license and use cases.
deepseek-coder-6.7b-instruct size and hardware requirements
6.7B
Total parameters
Dense (no MoE)
Architecture
BF16
Published precision
15.1 GB
Min VRAM (native)
| Precision | Weight size on disk | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 12.6 GB | 15.1 GB | RTX A4000 | 1 | $0.113/hr |
| FP8 (quantized) | 6.3 GB | 7.5 GB | RTX 4070 Super | 1 | $0.110/hr |
| INT4 (quantized) | 3.1 GB | 3.8 GB | RTX 4070 Super | 1 | $0.110/hr |
How to run deepseek-coder-6.7b-instruct
Run deepseek-coder-6.7b-instruct with vLLM
Generic example, not from the model's own docs: adjust flags (quantization, context length, parallelism) for your setup.
vllm serve deepseek-ai/deepseek-coder-6.7b-instruct --tensor-parallel-size 1Run deepseek-coder-6.7b-instruct with Ollama
Verified against Ollama's own library listing.
ollama run deepseek-coder:6.7bRun deepseek-coder-6.7b-instruct with GGUF quantizations
Prebuilt GGUF weights published at TheBloke/deepseek-coder-6.7B-instruct-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf TheBloke/deepseek-coder-6.7B-instruct-GGUFSource: https://huggingface.co/TheBloke/deepseek-coder-6.7B-instruct-GGUF
Deploy deepseek-coder-6.7b-instruct on Aquanode
Aquanode has no one-click deploy template for deepseek-coder-6.7b-instruct; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (1× RTX A4000 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU pod