DeepSeek-V3.1-Nex-N1 671B — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
DeepSeek V3.1 post-trained by Nex-AGI for agent autonomy and tool use — the base model's 671B weights at the base model's size, so 406 GB at Q4_K_M. Scores 80.2 on tau2-Bench, 70.6 on SWE-bench Verified, 65.3 on BFCL v4 and 29.5 on GAIA2. MLA attention keeps the KV cache small enough for a 128K context at this scale.
DeepSeek-V3.1-Nex-N1 671B needs about 406 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 222 GB | — | Too big | |
| Q3_K_M 3.41 bpw | 287.4 GB | — | Too big | |
| Q4_K_M 4.83 bpw | 406.5 GB | — | Too big | |
| Q5_K_M 5.67 bpw | 476.9 GB | — | Too big | |
| Q6_K 6.56 bpw | 551.6 GB | — | Too big | |
| Q8_0 8.50 bpw | 714.3 GB | — | Too big | |
| F16 16.00 bpw | 1343.4 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. DeepSeek-V3.1-Nex-N1 671B VRAM calculator →
Get DeepSeek-V3.1-Nex-N1 671B running
DeepSeek-V3.1-Nex-N1 671B is not practical on a single consumer GPU: it needs about 405.9 GB at Q4_K_M, more than any single device in our catalogue holds with headroom. How multi-GPU inference works, or rent a multi-GPU server by the hour:
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run DeepSeek-V3.1-Nex-N1 671B
Install Ollama, then run:
ollama run nex-n1Specifications
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
- Parameters
- 671 Billion (37B active)
- Context window
- 128,000
- Architecture
- Mixture-of-Experts (DeepSeek V3.1 post-train, MLA attention)
- Provider
- Nex-AGI
- Licence
- MIT
- Practical deployment
- Multi-GPU server
- Specified at
- Q4_K_M
- System RAM
- 512 GB
- Record updated
- 2026-09-08
Commercial use permitted. No usage restrictions beyond attribution.
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Can I run DeepSeek-V3.1-Nex-N1 671B on my GPU?
DeepSeek-V3.1-Nex-N1 671B — frequently asked questions
How much VRAM does DeepSeek-V3.1-Nex-N1 671B need?
About 406 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does DeepSeek-V3.1-Nex-N1 671B run on an RTX 4090 (24 GB)?
No. DeepSeek-V3.1-Nex-N1 671B needs about 406 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run DeepSeek-V3.1-Nex-N1 671B locally?
Install Ollama and run `ollama run nex-n1`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.