Models / Run locally
Run DeepSeek V4 Flash locally
DeepSeek · ≈40B-class compact tier · open weights
The one V4-family model within hobbyist reach: a 4-bit quant fits a single 24–32GB GPU (RTX 4090/5090-class) or a 36GB+ Mac with usable speed.
By quantization
Hardware requirements
| Quantization | ≈ File size | ≈ Memory needed | Runs on |
|---|---|---|---|
| Q4_K_M (4-bit) | 23 GB | 26 GB | 48GB workstation GPUs or 64GB Macs |
| Q8_0 (8-bit) | 43 GB | 49 GB | Multi-GPU workstations or 128–192GB Mac Studios |
| FP16 (full) | 84 GB | 97 GB | Multi-GPU workstations or 128–192GB Mac Studios |
Sizes are computed from the ≈40B parameter count with standard GGUF math and ~15% runtime overhead — verify against the model card before buying hardware.
Local setup
How to run it
Grab a community GGUF conversion from Hugging Face and serve it with llama.cpp, LM Studio, or (for multi-GPU rigs) vLLM with the original weights. Pick the quantization from the table above that fits your memory.
For serious throughput on server hardware, vLLM with tensor parallelism is the standard serving stack; llama.cpp and MLX (Apple silicon) win below that line.
The API route
Or skip the GPU entirely
The same model is one API call away at $0.15 / M input · $0.45 / M output — no download, no VRAM math, metered to the exact call. Free starter credits cover your first runs, and because the weights are open, nothing locks you in: start metered, move to your own hardware if sustained volume ever justifies it.
FAQ
Frequently asked questions
Can my GPU run DeepSeek V4 Flash?
DeepSeek V4 Flash is ≈40B-class compact tier. At 4-bit quantization you need roughly 26GB of memory (48gb workstation gpus or 64gb macs). The one V4-family model within hobbyist reach: a 4-bit quant fits a single 24–32GB GPU (RTX 4090/5090-class) or a 36GB+ Mac with usable speed.
Is there a DeepSeek V4 Flash GGUF?
Open-weight releases in this family get community GGUF conversions on Hugging Face shortly after release — search the model name plus "GGUF" and pick the quantization your memory allows from the table above.
Is running DeepSeek V4 Flash locally cheaper than the API?
Only at sustained volume. The API price is $0.15 / M input · $0.45 / M output with no hardware, electricity, or ops cost — a machine that can serve this model costs more per month idle than most teams' entire inference bill. Prototype metered, self-host when utilization justifies it.
Can I use the outputs commercially?
Open weights; commercial use permitted (MIT-style model license).
Explore
Run other models locally
One key. Every model. Exact prices.
Prototype on DeepSeek V4 Flash with free starter credits while the weights download. Free starter credits included — no subscription required.