Nex-N2.6
API 👉 OpenRouter
Nex-N2.6 is a multimodal agentic model from Nex-AGI, built on GLM-5.3-Flash. It is designed for computer use, coding, and long-horizon tasks involving tools. This repository contains the FP8-quantized weights of Nex-N2.6, with approximately 321B parameters.
Following Nex-N2.5-Pro, this release focuses on reducing repetitive reasoning and improving adherence to the tools available in the current environment.
Highlights
- 🔥 Less repetitive reasoning: Reasoning repetition fell from 16.0% to 1.6% compared with Nex-N2.5-Pro in our internal checkpoint-replay evaluation—a 90% relative reduction.
- 🛠️ Improved tool-list adherence: Nex-N2.6 achieved 100% tool-list adherence among responses making tool calls in our internal evaluation.
- 🖥️ Stronger computer use: Nex-N2.6 scored 83.0 on OSWorld, up from 82.2 for Nex-N2.5-Pro and 78.0 for the base model, GLM-5.3-Flash.
Performance
The following results are reported by the Nex-AGI team for this release. Scores depend on the evaluation harness and settings. These are model-level results, not a separate evaluation of this FP8 export.
| Benchmark ↑ | GLM-5.3-Flash | DeepSeek-V4.1-Flash | Nex-N2.5-Pro | Nex-N2.6 |
|---|---|---|---|---|
| OSWorld | 78.0 | 71.7 | 82.2 | 83.0 |
| Terminal-Bench 2.1 | 84.3 | 82.7 | 82.7 | 84.1 |
| AutomationBench | 48.8 | 54.8 | 44.2 | 45.7 |
Internal reliability evaluations
These evaluations are built from issues encountered in real tasks by our team and partner teams. Known-issue replay fixes the preceding conversation and tool outputs, then tests the model's next response. Tool-list adherence checks whether generated calls use only the tools declared in the request.
| Metric | Nex-N2.5-Pro | GLM-5.3-Flash | Nex-N2.6 |
|---|---|---|---|
| Reasoning Repetition Rate ↓ | 16.0% | 6.0% | 1.6% |
| Overall Non-Repetition Rate ↑ | 91.0% | 96.2% | 97.8% |
| Valid Tool Selection Rate (Overall) ↑ | 82.2% | 85.3% | 100.0% |
| Valid Tool Selection Rate (When with Tools) ↑ | 80.5% | 84.1% | 100.0% |
Usage
Use an inference runtime with GLM-5.3-Flash support. For self-hosting, follow the official SGLang deployment guide and select an FP8-compatible configuration for your hardware. Replace the model path with the downloaded Nex-N2.6 directory and set --served-model-name Nex-N2.6. Keep the tokenizer, processor, and chat template shipped with these weights.
Once the OpenAI-compatible server is listening on port 8000:
curl -N http://localhost:8000/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{
"model": "Nex-N2.6",
"messages": [
{"role": "user", "content": "Write a Python function that checks whether a string is a palindrome."}
],
"reasoning_effort": "max",
"max_tokens": 8192,
"stream": true
}'
The bundled chat template supports reasoning_effort values low, high, and max, defaulting to max. For image inputs, use image content blocks supported by your serving runtime. For tool use, supply the available tools with each request; the client or agent framework executes tool calls and returns their results.
Community
Join the Nex community to share feedback, discuss agent workflows, and connect with other developers.
| Feishu / Lark | |
|---|---|