On the plus side, at least an exact string match on the system prompt means I don't have to worry that if my legitimate Claude Code usage occasionally mentions the string "OpenClaw" I might get hard-to-debug intermittent failures
CODE
-
OpenAI President Greg Brockman on AI Models and Takeoff
By
–
I spoke at length with @OpenAI President @gdb about the company's double-down bet on text models, AI takeoff, Codex, infrastructure scaling, and plenty more.
— Alex Kantrowitz (@Kantrowitz) 5 avril 2026
Full episode below:
0:00 Introduction
4:06 Why OpenAI Pulled Back From Sora
11:24 The OpenAI Super App Plan
22:59… pic.twitter.com/LRKdXB8DGPI spoke at length with @OpenAI President @gdb about the company's double-down bet on text models, AI takeoff, Codex, infrastructure scaling, and plenty more. Full episode below: 0:00 Introduction 4:06 Why OpenAI Pulled Back From Sora 11:24 The OpenAI Super App Plan 22:59 OpenAI's Forthcoming “Spud” Model 28:13 OpenAI’s Automated AI Researcher Plan 31:12 AI Risk, Safety, and Takeoff 55:15 The Logic Behind OpenAI’s Compute Spending 1:03:24 Why So Many People Still Distrust AI
→ View original post on X — @ceobillionaire, 2026-04-05 17:34 UTC
-
Hidden Costs of Building and Maintaining Open Source Tools
By
–
You’re not factoring in the cost of your time building and maintaining the openclaw.
-
How to Design a Neural Network
By
–
How to Design a Neural Network
→ View original post on X — @deeplearn007, 2026-04-05 16:20 UTC
-
Qwen3-Coder-Next-REAP Models Released with Compression Options
By
–
As promised, Qwen3-Coder-Next-REAP are out! This will fit with full context in 48-62GB of VRAM the results are very solid. I need some MLX and GGUF bros to finish the job and get some compressssions out. – 20% huggingface.co/0xSero/qwen3-… – 30% huggingface.co/0xSero/qwen3-…
→ View original post on X — @clementdelangue, 2026-04-05 16:17 UTC
-
Fine-tuning Gemma on TPU v5 with Kinetic, Keras and JAX
By
–
Tutorial on fine tuning Gemma on TPU v5 using Kinetic + Keras + JAX. Easiest stack to fully leverage TPUs at scale. Jigyasa Grover ✨ (@jigyasa_grover) Here is a quick start script including the setup, technical details, and a candid look at where Kinetic excels versus its current limitations 🪡 github.com/jigyasa-grover/ki… — https://nitter.net/jigyasa_grover/status/2038707745520812099#m
-
Gemma-4-21B-REAP Released with Improved Reasoning Performance
By
–
As promised! Gemma-4-21B-REAP is out! Results are great it held up really well and actually gained accuracy on reasoning tasks. MLX & GGUF bros do you thing! This should fit on as little as 12GB of vram with some context, or 16GB with full context huggingface.co/0xSero/gemma-…
→ View original post on X — @deeplearn007, 2026-04-05 16:02 UTC
-
Switching from Claude to Local Gemma 4 on MacBook Pro
By
–
Cancelled calude code subscription for April month, just moved this setup – MacBook pro – Gemma 4 No internet, no API costs, no limits
→ View original post on X — @clementdelangue, 2026-04-05 15:35 UTC
-
Qwen 3.6 Plus Replaces Claude: Developer Switches After 90M Tokens
By
–
Its official. 90M tokens later. Qwen 3.6 Plus took all Opus tasks like a king! Since then my Claude usage has been 0. Not a single request! Not a single chat message. The best of it all, Qwen 3.6 Plus is behaving like SOTA models, with little to no errors on super complex, multi sub agents, multi-tasks, long-running tasks, you name it. You know what, I feel a little bad for Anthropic, it did help me build great tools such as truelens.tech that is changing the construction market in UK. If they did not do this super wrong move, I would still be there, so thank you @AnthropicAI I know there are 1000's of other amazing builders doing the exact same as we speak, but with other models. The point is that Anthropic jus did what it was not supposed to do, pushing those builders away, and once they run their own local models, those companies wont see a single cent anymore.
→ View original post on X — @clementdelangue, 2026-04-05 15:25 UTC
-
Coding agents outperform long-context models
By
–
BREAKING: Duke researchers just proved that coding agents are better at processing long documents than models with million-token context windows. > Not because of longer context. Because grep and sed are better retrieval tools than attention. > +17.3% average improvement