/Tech3h ago

Nex-AGI Unveils Nex-N2 Open-Weight Agentic Models With Strong Benchmarks

6450135.2K
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)#501
Original post
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex#501inTech

They claim they've pushed Qwen 3.5-397B to 33.6 on DeepSWE. This is well ahead of Kimi, DeepSeek and (as far as I know) all actual Qwens. I'm skeptical

Nex@NexEcosystem

📢 Nex-N2 is here! A family of agentic models that doesn't just think, it acts! Coding, search, tool use. All fused into a single agentic reasoning loop.

- Adaptive Thinking, auto-scales reasoning depth per step. Saves ~20% tokens, zero performance loss. - Coherent Thinking, one thinking paradigm across search, coding, and tool use. No more fragile mode-switching.

🏆 Result: Tier-1 open-source performance on SWE-bench, Terminal-Bench, GDPval, and more, tracking GPT-5.5 and Opus 4.7.

🎉 Open-weight. Try it now. 🔗 https://nex-agi.com/ 📦 https://huggingface.co/nex-agi/Nex-N2-Pro https://www.modelscope.cn/models/nex-agi/Nex-N2-Pro https://github.com/nex-agi/Nex-N2

12:01 PM · Jun 11, 2026 · 3.6K Views
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS1.6KLIKES4
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTex

makes sense some vintage china style

ρ:ɡeσn@pigeon__s

this model has insane claudey CoT except without the part about being short it overthinks like fucking crazy they are just lying here about the CoT efficiency its fucking terrible its literally not this good i tried it instantly dissapoints im afraid they trained on DeepSWE

3hViews 1.6KLikes 4Bookmarks 0
Tennyson@TennysonLabs

@teortaxesTex Whatever makes you happy honey

3hViews 9