Ligh7Fun

Присоединяйтесь по моей ссылке и пройдите проверку, чтобы получить API-кредиты. cavoti.com/sign-up?aff=gw…

@CommandCodeAI @TaLaTaQuangKhoi + zed

@TaLaTaQuangKhoi Quite soon. Where do you want it.

Hey @CommandCodeAI , do you support ACP?

@ShaoShur Или попробовать китайские модели, если постараться, получается не хуже, а то и лучше)

@ShaoShur Поискать готовые акки с подпиской, в среднем около 10$, пару раз брал(в тг), доживали до конца

Мне нафиг уже второй аккаунт заблокировали на чат GPT. У всех так?

@nstrumss А в подписке их около 80 + ещё сервера с б/с

@nstrumss Попробуйте геодему есть бесплатные серверы t.me/Geodema_bot?st…

Я очень хвалила впн которым пользуюсь, не лагал, но на этой неделе он стабильно выдает мне нерабочие серверы это как понимать блять

@CommandCodeAI I'm having some issues logging into my personal account today. I can't log in, and the model change process in the cmd is taking a really long time—it takes about 20-30 seconds for the model list to appear. Is this a common issue, or is it just me?

@CommandCodeAI @neeraj___ai I wasn't prepared for this; there were no tasks to solve this week, so I only managed to get 31m done 😀

@neeraj___ai Woohoo! 🥳🥳🥳 there's a share button with a nice image if you wanna share that.

It's so much worth for 1$ @CommandCodeAI

@PradyuPrasad @CommandCodeAI I recommend giving it a try—it's a great tool 👍

how are non-CC/Codex coding agent harnesses now? Do they mostly use Codex subscription or open source models? How do they compete with the big 2?

@Da7_Tech It seems to me that they don't have a cache, or they aren't taking it into account...

What a disaster. This genuinely goes against everything people expect from the values behind open-source models. Yesterday, Zcode was released — ZAI’s own app, built to work with their GLM models. I tried it on the Pro plan, the same Pro plan they describe as “15x Claude Code.” One single task consumed 60 million tokens from my limits in less than an hour, burning through the entire 5-hour limit without even completing the task. And the worst part? Their own app clearly showed the task used under 2 million tokens. But according to my tests, ZAI still deducted around 60 million tokens from both the 5-hour limit and the weekly limit. That strongly suggests cached tokens are being counted against me as real usage. So in practice, I am not even getting 5% of the value of my subscription. The same task was completed by Codex and MiniMax on their $20 plans. That says a lot about ZAI’s current infrastructure. I really hope they fix this, because I do want their models to succeed. I believed in the project enough to subscribe for a full year. But so far, for more than a month, I have not been able to actually benefit from my subscription. Let’s see what happens in the next few days.

@TheMaran Something very similar to agentrouter. The API isn't working; the response is a 504 Gateway Timeout. 👎

you can test gpt 5.5, deepseek v4 flash, glm 5.1, and more for $0 😳 this platform gives $100 free credits to test models. no need to give any card details what you get for free: - $100 worth credits - top models like chatgpt 5.5 - you can test 10+ models for free how to get credits: > go to: api.bluesminds.com - signup with github - go to the token management section & create token - save the token & get the base url how to use the token & base url: - you can just give the base url to grok - ask how to use the base url with the token - you need to tell grok where you want to use it specifically - like using in cursor, hermes agent, etc, etc. things you need to know: - is it enough to build projects? no - you can actually experience those models - except gpt 5.5, others are not that good - gpt 5.5 consumes a lot of tokens here - i suggest you to use an alt gitHub account bookmark this before paying for any big model api

@sahlbar Уезжать из России 😁

bayangin kalo misalnya: - x diblokir - whatsapp diblokir - instagram diblokir - tiktok diblokir - telegram diblokir - youtube diblokir kalian mau ngapain?

@CommandCodeAI Go GLM-5.2?

@florentfine Arch

be honest linux users. Fedora or Ubuntu?

@nalinrajput23 @zeddotdev

honestly , which IDE is the BEST right now? -Vs code -Antigravity -Cursor -Zed

ChatGPT 95% of the monthly limit remained before the request 1 request - create html+css landing Context window: 80% left (61.8K used / 258K) │ Monthly limit: [███████░░░░░░░░░░░░░] 34% left good job

@Freerunnering С кодом qwen справляется куда лучше

Also tested this was Qwen3.6 35B-A3B. It's a smarter model, but MTP doesn't give the same speed boost. Only around 56 tk/s, instead of Gemma's 70 tk/s.

This actually makes Gemma 4 26B-4A usable for a coding agent @ 72tk/s on my MacBook Pro M1 Max. This video is realtime, running completely locally.

@JoelDeTeves Qwen3.6-35b-a3b is better at writing human-like text, but you need a good prompt

I accidentally discovered that Gemma-4-26B-A4B is way better at writing human sounding content than every other model out there - including frontier models like GPT 5.5 and Sonnet 4.6. I'm not sure why this is - it's kind of crazy how slopified these big expensive models are and for some reason, Google's open source model sounds a lot more natural and follows writing instructions better. WTF?

@CodexReleases Сделать нормальные лимиты - нет. Сделать сброс лимитов - да!!! 🫩

Codex app update: Codex app 26.609 What changed: • Rate-limit reset banking added for Plus and Pro users, with one free reset at launch and referral invitations to earn more during the current promotion. • Developer mode for Browser use adds controlled CDP access in Chrome and the in-app browser for performance profiling and deeper debugging. • Browser use is up to 2x faster via CDP and DOM snapshot optimizations that reduce browser round trips. Details in thread.

@FM_yassin Вы сами то проверяли, как оно работает? 🤣

🚨 Most developers are paying separately for Claude, GPT, Gemini, and DeepSeek. Meanwhile, I just found an AI gateway that gives access to 30+ models with $125 in free credits and no credit card required. The project is called AgentRouter. What you get: → $125 free credits on signup → Claude Sonnet 4.8 & Opus 4.5 → GPT-4o & GPT-4o Mini → DeepSeek V3 & R1 → GLM-5.1 → Qwen 3 → Gemini → 25+ additional models The best part? One API key. One base URL. Works with: ✅ Claude Code ✅ Cursor ✅ Hermes ✅ Continue ✅ OpenAI-compatible tools Setup takes about 2 minutes: Sign in with GitHub Generate an API key Add the base URL Start using any model you want No juggling multiple accounts. No managing multiple API providers. Perfect for side projects, AI agents, prototypes, and experimentation. The AI tooling landscape is changing fast. Access to top-tier models is becoming cheaper every month. Part 2 👇 #AI #ClaudeCode #GPT4o #DeepSeek #OpenSource #GitHub #Developer #AIAgents #Automation #MachineLearning