Nex-N2-Pro
Nex AGI 于 2026 年 6 月发布的开源 Agent 模型。SWE-Bench Pro 得分 58.8,超越 GPT-5.5。397B 总参数 MoE 架构,仅 17B 激活,262K 上下文。独创 Adaptive Thinking 自适应推理 + Coherent Thinking 统一推理范式。Apache 2.0 完全开源,OpenRouter 免费 API 可用。限时免费!
Nex AGI's open-source Agent model released June 2026. SWE-Bench Pro score 58.8 — beating GPT-5.5. 397B MoE with only 17B active params, 262K context. Novel Adaptive Thinking + Coherent Thinking. Apache 2.0 fully open source, free API via OpenRouter. Limited-time free!
📊 基准测试对比Benchmark Comparison
| Benchmark | Nex-N2-Pro | GPT-5.5 | Claude Opus 4.7 |
|---|---|---|---|
| SWE-Bench Verified(编码) | 80.8 | 82.9 | 87.6 |
| SWE-Bench Pro(专业编码) | 58.8 ⬆超GPT-5.5 | 58.6 | 64.3 |
| Terminal-Bench 2.1(终端操作) | 75.3 ⬆超Opus | 83.4 | 69.7 |
| GPQA Diamond(研究生推理) | 90.7 | 93.6 | 94.2 |
| BrowseComp(浏览理解) | 83.7 | 84.4 | 79.8 |
| IFEval(指令遵循) | 94.0 | — | — |
| SWE-Bench Verified (Coding) | 80.8 | 82.9 | 87.6 |
| SWE-Bench Pro (Pro Coding) | 58.8 ⬆Beats GPT-5.5 | 58.6 | 64.3 |
| Terminal-Bench 2.1 (Terminal) | 75.3 ⬆Beats Opus | 83.4 | 69.7 |
| GPQA Diamond (Grad Reasoning) | 90.7 | 93.6 | 94.2 |
| BrowseComp (Browsing) | 83.7 | 84.4 | 79.8 |
| IFEval (Instruction Following) | 94.0 | — | — |
🧠 模型规格Model Specs
| 规格 | Spec | Nex-N2-Pro | Nex-N2-mini |
|---|---|---|---|
| 总参数 | 397B | 35B | |
| 每轮激活参数 | 17B (MoE) | 3B (MoE) | |
| 上下文窗口 | 262,144 tokens | 128K tokens | |
| 最大输出 | 256K tokens | 128K tokens | |
| 架构 | Transformer MoE | Transformer MoE | |
| 基座模型 | Qwen3.5-397B-A17B | Qwen3.5-35B-A3B | |
| 输入模态 | 文本 + 图像 | 文本 + 图像 | |
| 开源协议 | Apache 2.0 ✓ | Apache 2.0 ✓ | |
| 自部署显存 | ~794 GB BF16 / 216 GB INT4 | ~70 GB BF16 / 20 GB INT4 | |
| Total Params | 397B | 35B | |
| Active per Forward | 17B (MoE) | 3B (MoE) | |
| Context Window | 262,144 tokens | 128K tokens | |
| Max Output | 256K tokens | 128K tokens | |
| Architecture | Transformer MoE | Transformer MoE | |
| Base Model | Qwen3.5-397B-A17B | Qwen3.5-35B-A3B | |
| Input Modality | Text + Image | Text + Image | |
| License | Apache 2.0 ✓ | Apache 2.0 ✓ | |
| Self-Host VRAM | ~794 GB BF16 / 216 GB INT4 | ~70 GB BF16 / 20 GB INT4 |
🔬 核心技术Core Innovations
🚀 免费使用方式How to Access (Free)
❓ FAQ
Nex-N2-Pro 是免费的吗?
是的!通过 OpenRouter 和 SiliconFlow 提供的 API 完全免费。模型本身 Apache 2.0 开源,下载权重自行部署也完全免费。是限时免费活动,建议尽快体验。
和 GPT-5.5 比真的更强吗?
在 SWE-Bench Pro(专业编码)和 Terminal-Bench(终端操作)上确实超越了 GPT-5.5。在通用推理上略逊(GPQA Diamond 90.7 vs 93.6),但已经是目前最接近闭源旗舰的开源免费模型。
我能在自己电脑上跑吗?
Pro 版(397B)需要数据中心级硬件(~794GB 显存 BF16,或 ~216GB INT4)。个人用户建议用 OpenRouter 免费 API 或下载 mini 版(35B/3B,~70GB BF16 / ~20GB INT4,单张 RTX 4090+ 可跑)。
Is Nex-N2-Pro really free?
Yes! API access via OpenRouter and SiliconFlow is completely free. The model itself is Apache 2.0 open-source — download and self-host at no cost. Limited-time free offer, try it now.
Does it really beat GPT-5.5?
On SWE-Bench Pro (coding) and Terminal-Bench (terminal ops), yes. On general reasoning it's slightly behind (GPQA Diamond 90.7 vs 93.6), but it's the closest open-source free model to closed-source flagships.
Can I run it on my own machine?
Pro (397B) needs data-center hardware (~794GB VRAM BF16, ~216GB INT4). For personal use, try the free API or the mini variant (35B/3B, ~20GB INT4 — runs on RTX 4090+).