DeepSeek V3.1 — Hardware Requirements & GPU Compatibility
ChatFunctions
DeepSeek-V3.1 is a hybrid instruct model from DeepSeek that supports both thinking and non-thinking modes within a single set of weights, switched via its chat template. It is a roughly 684-billion-parameter Mixture-of-Experts model with 37 billion parameters activated per token, post-trained on top of DeepSeek-V3.1-Base with an extended long-context training phase, adding improved tool calling and dedicated formats for code and search agents. It has a 163,840-token context window and is released under the MIT license. Even at 4-bit quantization it needs roughly 394GB of memory, so local deployment requires server-class or multi-GPU hardware rather than a single machine.
DeepSeek V3.1 runs on a 512 GB+ Mac or a multi-GPU server
Best local pick → Mac Studio (M3 Ultra, 512GB) · Q4_K_M · 412.2 GB · ~18 tok/s
Macs from 256 GB (IQ2_M · ~28 tok/s)
Score 100.2#56 of 106 open#52 of 102 that fit ≤512 GBMIT
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
llmrun Composite Score →
100.2
- 90% range 95.4–105.0
- #56 of 106 open models
- #52 of 102 that fit ≤512 GB
Run DeepSeek V3.1
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve deepseek-ai/DeepSeek-V3.1 --max-model-len 16384 --tensor-parallel-size 8Size the nodes from the table above
Download: deepseek-ai_DeepSeek-V3.1-Q4_K_M/deepseek-ai_DeepSeek-V3.1-Q4_K_M-00001-of-00011.gguf409.23 GB + 10 partsbartowski
Ways to runOther ways
each opens step-by-stepRun in cloud
Rent 6× H100 SXM · Q4_K_M
- Estimated hourly price
- ~$16.14/hr
- est., whole server
- GPUs
- 6 GPUs
- 480 GB of VRAM
$5+ bonus for new users
How Much VRAM Does DeepSeek V3.1 Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 294.8 GB | 293.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Runs on 5× H100 SXM 80 GB | |
| 303.4 GB | 302.1 GB | 3-bit small — 3-bit small quantization | Runs on 5× H100 SXM 80 GB | |
| 337.6 GB | 336.3 GB | 3-bit medium — 3-bit medium quantization | Runs on 5× H100 SXM 80 GB | |
| 346.1 GB | 344.9 GB | 4-bit · legacy — 4-bit legacy quantization | Runs on 5× H100 SXM 80 GB | |
| 414.6 GB | 413.3 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Runs on 6× H100 SXM 80 GB | |
| 491.6 GB | 490.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Runs on 7× H100 SXM 80 GB | |
| 568.6 GB | 567.3 GB | 6-bit, very good — 6-bit quantization, very good quality | Runs on 8× H100 SXM 80 GB | |
| 688.4 GB | 687.1 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Runs on 6× H200 SXM 141 GB |
Selected Q4_K_M
4-bit medium quantization412.2GBat 16K context
- Weights
- 410.7 GB
- KV cache
- 1.2 GB
- Overhead
- 0.3 GB
Server-class: ~412.2 GB at 16K context · about 6× H100 SXM 80 GB
Which Hardware Can Run DeepSeek V3.1?
Q4_K_M · 414.6 GBDeepSeek V3.1 (Q4_K_M) requires 414.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 539+ GB is recommended. Using the full 164K context window can add up to 7.9 GB, bringing total usage to 422.5 GB. No consumer GPU has enough memory. 1 device with unified memory can run DeepSeek V3.1, including Mac Studio (M3 Ultra, 512GB).
192+ GB
Q4_K_M · best fit IQ2_XXS (189.7 GB)
4 GPUs · 1 Mac class · 2 device families
B300~181Simulate output on B300Instinct MI350X~69Simulate output on Instinct MI350XInstinct MI355X~69Simulate output on Instinct MI355XM3 Ultra256–512 GB~31Simulate output on Mac Studio (M3 Ultra, 256GB)8x H100 SXM5~250Simulate output on NVIDIA DGX H1008x A100 SXM4~226Simulate output on NVIDIA DGX A100 640GB
By VRAM tier · typical GPUs and Macs at Q4_K_M
Consumer GPUs (up to 32 GB) and Macs up to 192 GB: too small for this model (needs ≥ 189.7 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
How good is DeepSeek V3.1?
The llmrun Score merges 7 public benchmark boards into a single figure, where 100 represents the average for open models.
Score vs VRAM in the ≤512 GB tier
- DeepSeek V3.1
- Other models
- Best Score for its VRAM
#52 of 102 that fit ≤512 GB
Each dot is a model at the best quant that fits ≤512 GB. The line steps up to the best Score available at each amount of VRAM; a dot below it is beaten by a model that needs no more memory.
Full ≤512 GB ranking →Benchmarks
The 6 boards that weigh most in the llmrun Score, of 8 with a result for this model; badges show who ran the test.
- LMArena Text
LMArena · Oct 2026Arena
Arena1416.9
#37/188 open
- MMLU-Pro
TIGER-LabIndependent
Independent84.8%
#9/146 open
- SimpleBench
Epoch AI · Aug 2025Independent
Independent40.0%
#14/25 open
- WeirdML
Epoch AI · Aug 2025Independent
Independent38.4%
#28/47 open
- DTBench
Epoch AI · Aug 2025Lab-reported
Lab-reported82.7%
#19/70 open
- LMCA
Epoch AI · Aug 2025Lab-reported
Lab-reported24.3%
#31/56 open
In the ≤512 GB class
102 models fit ≤512 GBSame Score, only models that fit ≤512 GB at their best quant: this model's peers by memory need.
- #1
DeepSeek V4.1 Flash 1017DeepSeek
134.4
458.4 GB
- #2
MiMo V2.6 Pro RLXiaomi
133.3
499.3 GB
- #3
MiMo V2.6 Flash RLXiaomi
131.9
186.5 GB
- #4
GLM 5.3Z.ai
130.1
453.8 GB
- #5
DeepSeek V4.1 FlashDeepSeek
129.4
458.4 GB
- #6
Step 5 PreviewStepFun
128.0
407.5 GB
Alternatives to DeepSeek V3.1
DeepSeek V3 family · 9 models →Official models with an llmrun Score, each answering one question; sizes are each model's default quant (usually Q4_K_M) with a 16K-token context.
Best alternative at this size
DeepSeek V4.1 Flash 1017
DeepSeek
- 34.2 higher Score
- 46.3 GB more VRAM
Score 134.4 · 458.4 GB
Fits in less VRAM
GLM 5.3 Flash
Z.ai
- 23.9 higher Score
- 207 GB less VRAM
Score 124.1 · 205.2 GB
Other vendor
GLM 5.3
Z.ai
- 29.9 higher Score
- 41.6 GB more VRAM
Score 130.1 · 453.8 GB
Similar models
10| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| DeepSeek V4.1 FlashDeepSeek | 129.4 | 458.4 GB | —needs > 24 GB | similar score |
| Step 5 PreviewStepFun | 128.0 | 407.5 GB | —needs > 24 GB | similar score |
| GLM 5.2Z.ai | 122.9 | 453.8 GB | —needs > 24 GB | similar score |
| Beam 1.0 501B A23Breflection | 119.5 | 304 GB | —needs > 24 GB | similar score |
| GLM 5.1Z.ai | 117.8 | 454.1 GB | —needs > 24 GB | similar score |
| Qwen3.5 397B A17BAlibaba | 112.8 | 242.6 GB | —needs > 24 GB | similar score |
| GLM 4.7Z.ai | 109.1 | 217.9 GB | —needs > 24 GB | similar score |
| GLM 4.6Z.ai | 103.1 | 216.9 GB | —needs > 24 GB | similar score |
| GLM 4.5Z.ai | 100.9 | 217.9 GB | —needs > 24 GB | similar score |
| Qwen3 30B A3B Thinking 2507Alibaba | 100.7 | 19.4 GB | ~173 | similar score |
Specifications
- Publisher
- DeepSeek
- Family
- DeepSeek V3
- Parameters
- 684.5B
- Active parameters
- 40.1B per token
- Architecture
- DeepseekV3ForCausalLM
- Context Length
- 163,840 tokens
- Vocabulary Size
- 129,280
- Release Date
- 2025-08-21
- License
- MIT
Frequently Asked Questions
How much VRAM does DeepSeek V3.1 need?
DeepSeek V3.1 requires 414.6 GB of VRAM at Q4_K_M, or 1372.9 GB at BF16. Full 164K context adds up to 7.9 GB (422.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 684.5B × 4.8 bits ÷ 8 = 410.7 GB
KV Cache + Overhead ≈ 3.9 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 11.8 GB (at full 164K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run DeepSeek V3.1?
No — DeepSeek V3.1 requires at least 192.1 GB at IQ2_XXS, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
What's the best quantization for DeepSeek V3.1?
For DeepSeek V3.1, Q4_K_M (414.6 GB) offers the best balance of quality and VRAM usage. Q5_K_S (474.5 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 192.1 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run DeepSeek V3.1 on a Mac?
Yes — Mac Studio (M3 Ultra, 256GB) and 1 other Mac can run DeepSeek V3.1, the smallest at IQ2_M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 414.6 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio (M3 Ultra, 512GB).
Can I run DeepSeek V3.1 locally?
Yes — DeepSeek V3.1 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
What's the download size of DeepSeek V3.1?
At Q4_K_M, the download is about 410.72 GB. The full-precision BF16 version is 1369.06 GB. The smallest option (IQ2_XXS) is 188.25 GB.
Which GPUs can run DeepSeek V3.1?
No single consumer GPU has enough VRAM to run DeepSeek V3.1 at Q4_K_M (414.6 GB). Multi-GPU or professional hardware is required.
Which devices can run DeepSeek V3.1?
1 device with unified memory can run DeepSeek V3.1 at Q4_K_M (414.6 GB), including Mac Studio (M3 Ultra, 512GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.