Aditya Prasad Dhal
316 posts
👀
Kimi.ai@Kimi_Moonshot
Meet Kimi K2.6: Advancing Open-Source Coding 🔹Open-source SOTA on HLE w/ tools (54.0), SWE-Bench Pro (58.6), SWE-bench Multilingual (76.7), BrowseComp (83.2), Toolathlon (50.0), Charxiv w/ python(86.7), Math Vision w/ python (93.2) What's new: 🔹Long-horizon coding - 4,000+ tool calls, over 12 hours of continuous execution, with generalization across languages (Rust, Go, Python) and tasks (frontend, devops, perf optimization). 🔹Motion-rich frontend - Videos in hero sections, WebGL shaders, GSAP + Framer Motion, Three.js 3D. 🔹Agent Swarms, elevated - 300 parallel sub-agents × 4,000 steps per run (up from K2.5's 100 / 1,500). One prompt, 100+ files. 🔹Proactive Agents - K2.6 model powers OpenClaw, Hermes Agent, etc for 24/7 autonomous ops. 🔹Claw Groups (research preview) - bring your own agents, command your friends', bots & humans in the loop. - K2.6 is now live on kimi.com in chat mode and agent mode. For production-grade coding, pair K2.6 with Kimi Code: kimi.com/code - 🔗 API: platform.moonshot.ai 🔗 Tech blog: kimi.com/blog/kimi-k2-6 🔗 Weights & code: huggingface.co/moonshotai/Kim…
ART
@dhaladitya108 @fynnso Guess what, they did
Feels like ai rely though
English
was messing with the OpenAI base URL in Cursor and caught this
accounts/anysphere/models/kimi-k2p5-rl-0317-s515-fast
so composer 2 is just Kimi K2.5 with RL
at least rename the model ID
Cursor@cursor_ai
Composer 2 is now available in Cursor.
English
@ClementDelangue @leerob you caught your wife with other guy and she says "yeah i cheated", and you like:
English
Yep, Composer 2 started from an open-source base! We will do full pretraining in the future.
Only ~1/4 of the compute spent on the final model came from the base, the rest is from our training. This is why evals are very different.
And yes, we are following the license through our inference partner terms.
Fynn@fynnso
was messing with the OpenAI base URL in Cursor and caught this accounts/anysphere/models/kimi-k2p5-rl-0317-s515-fast so composer 2 is just Kimi K2.5 with RL at least rename the model ID
English
@itzadetunji1 @fynnso Actually, they can't do that. Kimi has a special modified MIT license for hyperscale commercial use. They've already reached over 20 million USD in monthly revenue, so it's important for them to display the name 'Kimi K2.5' on their products.
English
@fynnso Yup it is Kimi 2.5
It is opensource so that's why they can do that
English
T3 Code now supports Claude.
If you have the Claude Code CLI installed and signed in, you can use it with T3 Code.
Hopefully the lawyers won't make us remove this 🙃
English
@akaushik759 yeah true, I often find myself switching between Opus 4.6 and GPT 5.4 when one doesn't quite give me the results I'm looking for 😅
English
@dhaladitya108 I felt the same, opus however excels in both domains
English
Tried Codex today, man, Claude Code is so much better.
OpenAI is really bad at UI/UX and naming. At this stage and this level of money i’d expect a much more polished product, can say that their product team probably doesn’t use it as much.
English
Aditya Prasad Dhal retweetledi
Just hit 150+ followers🎉
To celebrate, I'm giving these 6 Hero Sections ( Fully Editable ) for FREE.
How to get :
• Like this post
• Comment "figma"
• Repost ( so others can have it too )
• Follow me and i'll dm you the file
Let's gooooo 🔥
English
I don't understand why people still use Codex.
Claude is at least 2x better IMHO.
English
Stop prompting AI to "make me a landing page."
200+ hours testing taught me this:
Generic prompts → pretty pages
My prompts → pages that sell
The pack includes:
→ 3D Landing Hero Section (Today)
→ Sentient 3D Core
→ Cyberpunk Volumetric Shaders
→ Microlender Protocol
→ Hyper-Futuristic MCP-2099
Like + reply "PROMPT-1" for the library.
(following required for DM)
English