Nex-N1 — local AI model by Nex-AGI
Written by Jakub Rusinowski · Last updated
Nex-AGI's first agentic series, and the one that set the pattern: take someone else's open base weights, post-train them for tool use and autonomy, publish the result. The flagship is a DeepSeek-V3.1 post-train scoring 80.2 on tau2-Bench and 70.6 on SWE-bench Verified. Superseded by Nex-N2, but kept because the question 'what does a 671B agentic post-train need?' has a factual answer that does not change when the model stops being recommended.
Variants
The smallest Nex-N1 variant needs about 406 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| DeepSeek-V3.1-Nex-N1 671B → 671B (37B active) | ~405.9 GB | 128,000 | ollama run nex-n1 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run Nex-N1 locally
Install Ollama, then pull the tag.
ollama run nex-n1Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: DeepSeek-V3.1-Nex-N1 671BCan I run Nex-N1 on my GPU?
Nex-N1 — frequently asked questions
How much VRAM does Nex-N1 need?
Nex-N1 needs about 406 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: DeepSeek-V3.1-Nex-N1 671B (406 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Nex-N1 on an RTX 4090 (24 GB)?
Nex-N1's smallest variant needs about 406 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for Nex-N1?
Q4_K_M is the best balance of quality and VRAM for Nex-N1 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Nex-N1 with Ollama?
Nex-N1 has no local Ollama tag — the published tag is cloud-hosted, so running it sends your prompts to a hosted GPU rather than your own machine.