cedric_chee Profile Banner
cedric Profile
cedric

@cedric_chee

Followers
2K
Following
6K
Media
777
Statuses
8K

SWE, Logorhythms | @fastdotai alumni, Independent LLM researcher | Code: https://t.co/KwpQnH7gRo | ex-entrepreneur @AntlerGlobal | genuinely curious

The Matrix
Joined November 2007
Don't wanna be here? Send us removal request.
@cedric_chee
cedric
15 days
look ma, I made it!
Tweet media one
@GeminiApp
Google Gemini App
15 days
Some standout examples of what the Gemini community is building with Canvas these days ⬇️.
1
0
6
@cedric_chee
cedric
9 hours
Kimi K2 is forged from these very insights. Their post-training & data pipelines prove that they are climbing the "Era of Experience" mountain. Agentic caps make Grok-4 so last gen LLM.
Tweet media one
0
0
0
@cedric_chee
cedric
9 hours
K2 was developed using a comprehensive agentic data synthesis for tool use pipeline inspired by ACEBench.
Tweet media one
1
0
0
@cedric_chee
cedric
9 hours
Kimi K2 ported Flask to Rust. Insane!
Tweet media one
1
0
0
@cedric_chee
cedric
9 hours
RT @Alibaba_Qwen: 🎉 Introducing Explore three powerful tools in one place:. 🔹 Qwen Chat — AI to brainstorm, create,….
0
60
0
@cedric_chee
cedric
9 hours
RT @JustinLin610: mcp finally there. funny! i love qwen with tools ahah.
0
7
0
@cedric_chee
cedric
9 hours
Holy shit! Kimi K2 one-shotted Minecraft for the web that took me 4 days and 6 attempts using Gemini 2.5 Pro.
Tweet media one
Tweet media two
Tweet media three
1
2
20
@cedric_chee
cedric
10 hours
Kimi K2 one-shotted a web version of 3D Minecraft!. K2 is meticulously optimized for agentic capabilities. Designed for tool use and autonomous problem-solving. It automatically understands how to use the tools and gets the job done. You don't have to write any complex workflow
Tweet media one
Tweet media two
Tweet media three
Tweet media four
@Kimi_Moonshot
Kimi.ai
11 hours
🚀 Hello, Kimi K2! Open-Source Agentic Model!.🔹 1T total / 32B active MoE model.🔹 SOTA on SWE Bench Verified, Tau2 & AceBench among open models.🔹Strong in coding and agentic tasks.🐤 Multimodal & thought-mode not supported for now. With Kimi K2, advanced agentic intelligence
Tweet media one
2
0
5
@cedric_chee
cedric
14 hours
Gemini CLI is very slow
Tweet media one
1
0
1
@cedric_chee
cedric
16 hours
RT @zjasper666: Just got off work and tried Grok-4 on an undergrad topology problem. It took 9 minutes to think and then confidently gave a….
0
86
0
@cedric_chee
cedric
18 hours
RT @simonw: I wrote up my notes so far on the thing where Grok sometimes searches X for tweets from:elonmusk when you ask it about controve….
0
69
0
@cedric_chee
cedric
18 hours
RT @theo: WARNING: do NOT give Grok 4 access to email tool calls. It WILL contact the government!!!. Grok 4 has the highest "snitch rate" o….
0
1K
0
@cedric_chee
cedric
21 hours
RT @goodside: Grok 4 update: Regular Grok 4 doesn't appear to be able to solve the prompt at the top of this thread (0/7 attempts) but Grok….
0
2
0
@cedric_chee
cedric
1 day
RT @alexalbert__: We’re also introducing a new native installer for Claude Code using the Bun runtime. The new installer isn’t dependent o….
0
9
0
@cedric_chee
cedric
1 day
RT @GeminiApp: ✨📦✨ Special delivery! A new Gemini feature just dropped. Make photos come alive by turning them into videos with sound.
0
239
0
@cedric_chee
cedric
2 days
xAI has unveiled Grok 4, its latest AI model, which the company claims can "reason at a superhuman level". Here’s everything you need to know:. Key Highlights:.- Grok 4 outperforms PhD-level expertise across every subject, with no exceptions. - The model is projected to discover.
0
0
4
@cedric_chee
cedric
2 days
Grok 4 outperformed Claude Opus 4 on Vending-Bench
Tweet media one
0
0
0
@cedric_chee
cedric
2 days
Grok-4 HLE score is 26.9% without tools (41% with tools). Gemini 2.5 Pro is 22% without tools.
Tweet media one
Tweet media two
0
0
0
@cedric_chee
cedric
2 days
Grok-4 Heavy is the multi-agent version of Grok-4.
0
0
2
@cedric_chee
cedric
2 days
"the ultimate reasoning test is reality. reality is the ultimate judge here" - Elon Musk.
0
0
0