LLMs · DeepSeek
DeepSeek LLM 7B Chat
- Parameters
- 6.9B
- Architecture
- Dense
- Max context
- 4K tokens
- License
- deepseek
- Released
- 2023-11
How to run it
Quickest · Ollama
One command downloads a ready-to-run version and starts a chat.
ollama run deepseek-llm:7bQuantized files · GGUF
Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.
huggingface.co/mradermacher/deepseek-llm-7b-chat-i1-GGUF (opens in a new tab)
ollama run hf.co/mradermacher/deepseek-llm-7b-chat-i1-GGUFOriginal weights · Hugging Face
Full-precision files from DeepSeek, for fine-tuning, vLLM, Transformers.
huggingface.co/deepseek-ai/deepseek-llm-7b-chat (opens in a new tab)
Memory needed
Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.
| Quantization | Quality | Weights | Total at 4K context | Download |
|---|---|---|---|---|
Q2_K |
Low | 2.8 GB | 5.4 GB | GGUF (opens in a new tab) |
Q3_K_M |
Low | 3.4 GB | 6.0 GB | GGUF (opens in a new tab) |
Q4_K_M |
Good | 4.2 GB | 6.8 GB | GGUF (opens in a new tab) |
Q5_K_M |
High | 4.9 GB | 7.5 GB | GGUF (opens in a new tab) |
Q6_K |
High | 5.7 GB | 8.3 GB | GGUF (opens in a new tab) |
Q8_0 |
Very high | 7.3 GB | 9.9 GB | |
F16 |
Full | 13.8 GB | 16.4 GB | Original weights (opens in a new tab) |
Best hardware for DeepSeek LLM 7B Chat
The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A380 (launch price $139) at Q2_K.
| Grade · points | Hardware | Quant | Memory | Speed | Fit |
|---|---|---|---|---|---|
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | AMD Radeon RX 9070 XT | Q8_0 | 10.7 GB / 16.0 GB | 49.8 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4090 | F16 | 17.6 GB / 24.0 GB | 42.0 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4070 Ti SUPER | Q8_0 | 10.7 GB / 16.0 GB | 60.8 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4080 SUPER | Q8_0 | 10.7 GB / 16.0 GB | 66.6 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 5060 Ti 16GB | Q8_0 | 10.7 GB / 16.0 GB | 40.5 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 5070 Ti | Q8_0 | 10.7 GB / 16.0 GB | 81.0 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 5080 | Q8_0 | 10.7 GB / 16.0 GB | 86.8 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Intel Arc B580 | Q6_K | 8.9 GB / 12.0 GB | 42.9 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 3060 12GB | Q6_K | 8.9 GB / 12.0 GB | 44.4 tok/s | Runs well |
| 63 ptsGood (B) · 63 of 100 points: Model quality 8.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4070 SUPER | Q6_K | 8.9 GB / 12.0 GB | 62.1 tok/s | Runs well |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.