DeepSeek Coder v2 Instruct — Hardware Requirements & GPU Compatibility
ChatCode
DeepSeek-Coder-V2-Instruct is DeepSeek's mixture-of-experts code and math model, continuing pretraining from an intermediate DeepSeek-V2 checkpoint with an additional 6 trillion tokens focused on code. It has about 235.7 billion total parameters with roughly 21.4 billion active per token, and DeepSeek reports coding and mathematical reasoning performance comparable to GPT-4 Turbo, expanding supported programming languages from 86 to 338 versus the original DeepSeek-Coder line. It is an instruction-tuned chat model, not a base model, intended for code generation, completion, and reasoning tasks. Given its total size, DeepSeek recommends an eight-GPU, 80GB-class server for BF16 inference, so it needs a substantial multi-GPU setup even once quantized. Context length is 163,840 tokens; DeepSeek's own materials describe this as an extension "from 16K to 128K" over the original Deepseek Coder. Model weights are released under DeepSeek's custom model license, which permits commercial use, while the surrounding code is MIT-licensed. It was published in June 2024.
11.8K downloads
701 likesSpecifications
- Publisher
- DeepSeek
- Family
- DeepSeek Coder
- Parameters
- 235.7B
- Architecture
- DeepseekV2ForCausalLM
- Context Length
- 163,840 tokens
- Vocabulary Size
- 102,400
- Release Date
- 2024-06-14
- License
- Other
Get Started
HuggingFace
How Much VRAM Does DeepSeek Coder v2 Instruct Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| IQ2_XS | 2.40 | 73.5 GB | 272.4 GB | 70.72 GB | Importance-weighted 2-bit, extra small |
| Q2_K | 3.40 | 103.0 GB | 301.8 GB | 100.19 GB | 2-bit quantization with K-quant improvements |
| Q3_K_M | 3.90 | 117.7 GB | 316.6 GB | 114.92 GB | 3-bit medium quantization |
| Q4_K_M | 4.80 | 144.3 GB | 343.1 GB | 141.44 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 170.8 GB | 369.6 GB | 167.97 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 197.3 GB | 396.1 GB | 194.49 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 238.6 GB | 437.4 GB | 235.74 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 474.3 GB | 673.1 GB | 471.48 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run DeepSeek Coder v2 Instruct?
Q4_K_M · 144.3 GBDeepSeek Coder v2 Instruct (Q4_K_M) requires 144.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 188+ GB is recommended. Using the full 164K context window can add up to 198.8 GB, bringing total usage to 343.1 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run DeepSeek Coder v2 Instruct?
Q4_K_M · 144.3 GB6 devices with unified memory can run DeepSeek Coder v2 Instruct, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio (M3 Ultra, 256GB).
Runs great
— Plenty of headroom
Decent
— Enough memory, may be tightWhere to Download DeepSeek Coder v2 Instruct
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Benchmarks
Benchmark details →Related Models
Frequently Asked Questions
- How much VRAM does DeepSeek Coder v2 Instruct need?
DeepSeek Coder v2 Instruct requires 144.3 GB of VRAM at Q4_K_M, or 474.3 GB at BF16. Full 164K context adds up to 198.8 GB (343.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 235.7B × 4.8 bits ÷ 8 = 141.4 GB
KV Cache + Overhead ≈ 2.9 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 201.7 GB (at full 164K context)
VRAM usage by quantization
Q4_K_M144.3 GBQ4_K_M + full context343.1 GB- Can NVIDIA GeForce RTX 5090 run DeepSeek Coder v2 Instruct?
No — DeepSeek Coder v2 Instruct requires at least 73.5 GB at IQ2_XS, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- What's the best quantization for DeepSeek Coder v2 Instruct?
For DeepSeek Coder v2 Instruct, Q4_K_M (144.3 GB) offers the best balance of quality and VRAM usage. Q5_K_M (170.8 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XS at 73.5 GB.
VRAM requirement by quantization
IQ2_XS73.5 GBQ3_K_M117.7 GBQ4_K_M ★144.3 GBQ5_K_M170.8 GBQ6_K197.3 GBBF16474.3 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run DeepSeek Coder v2 Instruct on a Mac?
DeepSeek Coder v2 Instruct requires at least 73.5 GB at IQ2_XS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run DeepSeek Coder v2 Instruct locally?
Yes — DeepSeek Coder v2 Instruct can run locally on consumer hardware. At Q4_K_M quantization it needs 144.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is DeepSeek Coder v2 Instruct?
At Q4_K_M, DeepSeek Coder v2 Instruct can reach ~68 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 144.3 × 0.65 = ~172 tok/s
Estimated speed at Q4_K_M (144.3 GB)
~172 tok/s~172 tok/s~68 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of DeepSeek Coder v2 Instruct?
At Q4_K_M, the download is about 141.44 GB. The full-precision BF16 version is 471.48 GB. The smallest option (IQ2_XS) is 70.72 GB.
- Which GPUs can run DeepSeek Coder v2 Instruct?
No single consumer GPU has enough VRAM to run DeepSeek Coder v2 Instruct at Q4_K_M (144.3 GB). Multi-GPU or professional hardware is required.
- Which devices can run DeepSeek Coder v2 Instruct?
6 devices with unified memory can run DeepSeek Coder v2 Instruct at Q4_K_M (144.3 GB), including Mac Pro M2 Ultra (192 GB), Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M2 Ultra (192 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.