DeepSeek V3.1 Terminus — Hardware Requirements & GPU Compatibility
Chat
DeepSeek-V3.1-Terminus is an updated checkpoint of DeepSeek-V3.1, a Mixture-of-Experts chat and reasoning model with roughly 40.1 billion active parameters out of about 684.5 billion total, fine-tuned from DeepSeek-V3.1-Base. This revision keeps the same capabilities as V3.1 while fixing issues reported by users, chiefly reducing mixed Chinese-English text and stray characters in output and further improving the model's Code Agent and Search Agent performance, including gains on BrowseComp, SWE-bench Verified, and Terminal-bench. Its scale requires a multi-GPU server; it is not something that fits on consumer hardware. Context length is 163,840 tokens. It is released under the MIT license, permitting unrestricted commercial and research use. It was published in September 2025, as a refinement of DeepSeek-V3.1 rather than a new base model.
24.7K downloads
369 likesSpecifications
- Publisher
- DeepSeek
- Family
- DeepSeek V3
- Parameters
- 684.5B
- Architecture
- DeepseekV3ForCausalLM
- Context Length
- 163,840 tokens
- Vocabulary Size
- 129,280
- Release Date
- 2025-09-22
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does DeepSeek V3.1 Terminus Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 294.8 GB | 577.8 GB | 290.93 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 303.4 GB | 586.3 GB | 299.48 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 337.6 GB | 620.6 GB | 333.71 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 346.1 GB | 629.1 GB | 342.27 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 414.6 GB | 697.6 GB | 410.72 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 491.6 GB | 774.6 GB | 487.73 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 568.6 GB | 851.6 GB | 564.74 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 688.4 GB | 971.4 GB | 684.53 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run DeepSeek V3.1 Terminus?
Q4_K_M · 414.6 GBDeepSeek V3.1 Terminus (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 283.0 GB, bringing total usage to 697.6 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run DeepSeek V3.1 Terminus?
Q4_K_M · 414.6 GB2 devices with unified memory can run DeepSeek V3.1 Terminus, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomWhere to Download DeepSeek V3.1 Terminus
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 V3.1 Terminus need?
DeepSeek V3.1 Terminus requires 414.6 GB of VRAM at Q4_K_M, or 1372.9 GB at BF16. Full 164K context adds up to 283.0 GB (697.6 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)
KV Cache + Overhead ≈ 286.9 GB (at full 164K context)
VRAM usage by quantization
Q4_K_M414.6 GBQ4_K_M + full context697.6 GB- Can NVIDIA GeForce RTX 5090 run DeepSeek V3.1 Terminus?
No — DeepSeek V3.1 Terminus 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 Terminus?
For DeepSeek V3.1 Terminus, 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
IQ2_XXS192.1 GBIQ3_XS286.3 GBQ3_K_L354.7 GBQ4_K_M ★414.6 GBQ5_K_S474.5 GBBF161372.9 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run DeepSeek V3.1 Terminus on a Mac?
DeepSeek V3.1 Terminus requires at least 192.1 GB at IQ2_XXS, 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 V3.1 Terminus locally?
Yes — DeepSeek V3.1 Terminus can run locally on consumer hardware. At Q4_K_M quantization it needs 414.6 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of DeepSeek V3.1 Terminus?
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 Terminus?
No single consumer GPU has enough VRAM to run DeepSeek V3.1 Terminus at Q4_K_M (414.6 GB). Multi-GPU or professional hardware is required.
- Which devices can run DeepSeek V3.1 Terminus?
3 devices with unified memory can run DeepSeek V3.1 Terminus at Q4_K_M (414.6 GB), including Mac Studio (M3 Ultra, 512GB), NVIDIA DGX A100 640GB, NVIDIA DGX H100. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.