--- license: apache-2.0 pipeline_tag: text-generation library_name: transformers ---
---

๐Ÿ’ป GitHub  ยท   ๐Ÿค— Hugging Face  ยท   ๐ŸŒ Website  ยท   ๐Ÿ”€ OpenRouter (Pro)

๐Ÿšง  Nex-N2.5-Pro weights are coming soon.
# Nex-N2.5 **A next-generation family of agentic models built for long-horizon tasks in real-world environments.** Today, Nex-AGI officially introduces **Nex-N2.5**, its next-generation family of agentic models. Nex-N2.5 is available in three sizes: **mini**, **Pro**, and **Max**. Nex-N2.5-mini and Nex-N2.5-Pro continue to build on the multimodal foundations of Nex-N2, with focused improvements in computer use, web browsing, and visually grounded agentic capabilities. Nex-N2.5-Max is built on a 1.6-trillion-parameter, text-only Mixture-of-Experts (MoE) foundation model, marking our first complete post-training effort at trillion-parameter scale. For long-horizon tasks in real-world environments, Nex-N2.5 further strengthens its ability to act continuously and self-correct through visual feedback. The models can operate computers and browsers, as well as autonomously execute and test programs. Vision is therefore no longer merely an input modality; it has become a critical interface through which an agent perceives its environment, verifies outcomes, and moves a task forward. Building on this foundation, we have further expanded the range of agent training environments, task types, and productivity scenarios, while completing systematic post-training at trillion-parameter scale for the first time. Through broader task coverage and richer environmental feedback, Nex-N2.5 delivers further gains in scientific research, knowledge work, and complex productivity tasks. This work also provides valuable practical experience for training agentic capabilities in even larger models. By jointly advancing model training, infrastructure, and real-world agent scenarios, Nex-AGI aims to continue driving progress in agentic intelligence. ## Open Source Model weights for the Nex-N2.5 family will be released as open source, alongside hosted online services. - **Nex-N2.5-Max:** [Hugging Face](https://huggingface.co/nex-agi/Nex-N2.5-Max) | [ModelScope](https://modelscope.cn/models/nex-agi/Nex-N2.5-Max) - **Nex-N2.5-Pro:** [Hugging Face](https://huggingface.co/nex-agi/Nex-N2.5-Pro) | [ModelScope](https://modelscope.cn/models/nex-agi/Nex-N2.5-Pro) - **Nex-N2.5-mini:** [Hugging Face](https://huggingface.co/nex-agi/Nex-N2.5-mini) | [ModelScope](https://modelscope.cn/models/nex-agi/Nex-N2.5-mini) - **Hosted Access:** [OpenRouter (Nex-N2.5-Pro)](https://openrouter.ai/nex-agi/nex-n2.5-pro) - **Websites:** [Global](https://nex-agi.com/) We welcome developers and enterprises to integrate and try Nex-N2.5 and share their feedback. ## Performance We evaluate Nex-N2.5 across coding, agentic workflows, computer use, and multimodal understanding. ![Nex-N2.5 Benchmark Overview: Text and Multimodal](./figures/Nex-N2.5-Benchmark-white.png) The tables below compare **Nex-N2.5-mini**, **Nex-N2.5-Pro**, and **Nex-N2.5-Max** with leading models across our evaluation suite. **Bold** marks the best result in each benchmark, including ties; โ€” indicates unavailable data. ### Text Benchmarks
Benchmark Nex-N2.5-mini Nex-N2.5-Pro Nex-N2.5-Max Claude Opus 5 GPT-5.6 Sol Kimi-K3 GLM-5.3 DeepSeek-V4-Pro Qwen3.8-Max
CODING
Terminal-Bench 2.173.482.786.189.188.888.388.287.986.6
SWE-Bench Pro43.861.265.779.264.663.364.655.467.7
DeepSWE v1.136.155.865.673.772.767.566.962.869.3
AGENTIC
AutomationBench v1.0.632.344.250.250.345.846.748.243.250.8
Toolathlon Verified54.668.574.776.574.976.573.074.173.3
GDPval-AA v2144616281713186117481668176915541630
Job Bench28.541.453.665.745.452.958.254.164.0
BrowseComp83.489.792.690.890.491.2โ€”โ€”โ€”
### Multimodal Benchmarks
Benchmark Nex-N2.5-mini Nex-N2.5-Pro MiniMax-M3 Claude Opus 5 GPT-5.6 Sol Kimi-K3 GLM-5.3-Flash DeepSeek-V4-Flash-Vision Qwen3.8-Max
OSWorld-Verified71.282.275.283.483.284.862.376.786.1
OSWorld-230.556.422.368.362.758.3โ€”โ€”46.7
WebTest48.652.8โ€”โ€”54.0โ€”โ€”โ€”52.3
WebArena-Verified63.467.6โ€”โ€”69.771.6โ€”62.366.8
OSWorld-G82.987.4โ€”76.877.779.683.359.484.9
Vision2Web52.968.259.0โ€”79.8โ€”โ€”โ€”75.1
SWE-MM25.538.2โ€”59.440.237.320.639.239.2
OmniDoc89.792.291.6โ€”92.991.1โ€”โ€”92.1
## Usage ### Docker Deployment We also provide a prebuilt Docker image with our customized `sglang` fork preinstalled: **`nexagi/sglang:v0.5.18-nex-patch`**. The launch command is the same as above. #### Nex-N2.5-Max ```bash # Multi-node (2 nodes, 16 x H200). Run the same command on every node with: # = 0 on the head node, 1 on the other node # = IP of the head node (reachable from all others) docker run --gpus all --shm-size 32g --network host \ -v /path/to/your/model:/model \ nexagi/sglang:v0.5.18-nex-patch \ python3 -m sglang.launch_server \ --model-path /path/to/your/model \ --trust-remote-code \ --host 0.0.0.0 \ --port 8000 \ --nnodes 2 \ --node-rank "${NODE_RANK}" \ --dist-init-addr "${MASTER_ADDR}:5000" \ --tp 16 \ --pp-size 1 \ --dp 1 \ --ep-size 16 \ --attention-backend dsv4 \ --kv-cache-dtype fp8_e4m3 \ --page-size 256 \ --moe-a2a-backend deepep \ --moe-runner-backend deep_gemm \ --moe-dense-tp-size 1 \ --deepep-mode auto \ --context-length 262144 \ --mem-fraction-static 0.84 \ --chunked-prefill-size 8192 \ --enable-mixed-chunk \ --disable-overlap-schedule \ --max-running-requests 64 \ --cuda-graph-max-bs-decode 64 \ --cuda-graph-backend-decode full \ --cuda-graph-backend-prefill disabled \ --chat-template /path/to/nex-n2.5-max/chat_template.jinja \ --reasoning-parser deepseek-r1 \ --tool-call-parser qwen3_coder ``` #### Nex-N2.5-Pro Single node with 8 ร— H100: ```bash docker run --gpus all --shm-size 32g --ipc=host \ -p 30000:30000 \ -v /path/to/your/model:/model \ nexagi/sglang:v0.5.18-nex-patch \ python3 -m sglang.launch_server \ --model-path /model \ --tp 8 \ --host 0.0.0.0 --port 30000 \ --reasoning-parser qwen3 \ --tool-call-parser qwen3_coder \ --chat-template /path/to/nex-N2.5-Pro/chat-template.jinja \ --mamba-scheduler-strategy extra_buffer ``` #### Nex-N2.5-mini Single node with 2 ร— H100: ```bash docker run --gpus all --shm-size 32g --ipc=host \ -p 30000:30000 \ -v /path/to/your/model:/model \ nexagi/sglang:v0.5.18-nex-patch \ python3 -m sglang.launch_server \ --model-path /model \ --tp 2 \ --host 0.0.0.0 --port 30000 \ --reasoning-parser qwen3 \ --tool-call-parser qwen3_coder \ --chat-template /path/to/nex-N2.5-mini/chat-template.jinja \ --mamba-scheduler-strategy extra_buffer ``` ### Recommended Sampling Parameters For the best generation quality, we recommend the following sampling parameters: - `temperature`: 0.7 - `top_p`: 0.95 - `top_k`: 40 ### Function Calling Nex-series models support robust function-calling capabilities. To enable function calling, add the `--tool-call-parser qwen3_coder` flag when launching the server: ```bash python -m sglang.launch_server --model-path /path/to/your/model --tool-call-parser qwen3_coder ``` ### Reasoning Parser Nex-series models emit explicit reasoning traces. Add the `--reasoning-parser qwen3` flag to parse the reasoning content separately from the final response. It can be combined with the function-calling parser above: ```bash python -m sglang.launch_server --model-path /path/to/your/model --tool-call-parser qwen3_coder --reasoning-parser qwen3 ``` ### Thinking Control Nex-N2.5 models support three thinking modes, selected per request with the OpenAI-compatible `reasoning_effort` field: | `reasoning_effort` | Mode | Behavior | |---|---|---| | `"none"` | Thinking off | The model answers directly, without a reasoning trace. | | `"medium"` (default) | Adaptive thinking | The model decides per request whether, and how much, to reason. | | `"high"` | Thinking on | The model always produces a full reasoning trace before answering. | ```bash curl http://localhost:30000/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "nex-agi/Nex-N2.5-Pro", "messages": [{"role": "user", "content": "How many prime numbers are there below 100?"}], "reasoning_effort": "high" }' ```