@vinitpaul6 I am using the $20 plan. For heavy use i can no longer recommend the ollama plan. If you use opencode, getting the $5 promo plan is way more value for the money.
@ElnurIb01717668 i see my session usage increase by about 5% at times based on context, i've never seen this happen with glm 5.1 back then either, even k2.7 code eats up the usage. they definitely nerfed something.
@internetfreedom It's quite cluttered to go look for each particular topic. It would be great if you guys, along with pointing out issues like the Telegram ban, tweet or mention what you have done against such measures.
Recent posts on X have led to speculation regarding Reliance Jio Infocomm Limited (AS55836) and a BGP route misconfiguration. We categorically clarify that Jio has not been involved in any such incident. Jio continues to operate its network in accordance with global Internet routing best practices and the highest standards of reliability, security, and transparency.
Honestly the best instruction following! No other models other than GPT and Claude performs as good as this one in terms of following the instructions provided to it.
📢 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://t.co/7oLSfyOCxB
📦 https://t.co/c2CGhXWaz6
https://t.co/KJYXZIpk8M
https://t.co/vcjdZ9cuB6
For the love of god, none of GLM, Kimi, or any other Chinese models I’ve tested reliably follow my AGENTS.md instructions.
GPT/Claude treat the file like actual operating rules. The other models often treat it like loose context and just start editing files anyway.
Makes me wonder how close they really are to frontier level agentic behavior.
Dear people at @OpenAI. Could you add a toggle to run memory off easily in an individual chat? Sometimes I need a response not polluted by my previous conversations.
@nahcrof people using opensource models do tend to look for value and right now this ain't much value for money compared to other options right there...