Distractosphere retweeted
Try Underdog: free, capable, private AI running locally on your device
New build and Saluki 27B are out. Under 8GB, beats Qwen 3.8 27B at tool calling
Like, repost, and comment one task you wouldn’t give a cloud agent to get an invite🦴
Today we're releasing Underdog Saluki 27B
Based on Qwen 3.8 27B, Saluki is almost 7x smaller than its full-precision counterpart while it keeps 96% of its benchmark performance and beats Qwen at tool calling
It's the best model <8GB that runs in Underdog
Saluki is available today under Apache 2.0. Try today in Underdog underdog.ai/saluki
Distractosphere retweeted
To add some clarity: you don't pay extra. It's the same subscription, same price per month.
What's new our sub now covers two separate pools:
· Interactive → sub limits, unchanged
· Programmatic → new $20–$200 included(!!) credit, metered at API rates
Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage.
The credit covers usage of:
- Claude Agent SDK
- claude -p
- Claude Code GitHub Actions
- Third-party apps built on the Agent SDK
Readers added context they thought people might want to know
This is missing key context: programmatic use no longer draws from subscription limits. It uses a separate $20-$200 monthly Agent SDK credit, then extra usage is billed at standard API rates, so heavy sub users may get far less usage than before.
support.claude.com/en/articles/15…
platform.claude.com/docs/en/about-…
support.claude.com/en/articles/12…
Distractosphere retweeted
Computer use models shouldn't learn from screenshots.
We built a new foundation model that learns from video like humans do. FDM-1 can construct a gear in Blender, find software bugs, and even drive a real car through San Francisco using arrow keys.
Distractosphere retweeted
A bit sad how the GPT-5 launch is going so far, especially after the long wait and high expectations
- The automatic switching between models (the router) seems partly broken/unreliable
- It's unclear exactly which model you're actually interacting with (standard or mini, reasoning level, old/new)
- Router issues probably contribute to many negative first impressions ("GPT-5 is bad"), especially when you can't easily tell which model is handling your requests - e.g. GPT-5 Thinking performs better at writing compared to standard GPT-5 ("think harder" prompts might also help)
- I'm seeing lots of complaints about legacy models being removed and usage limits for Plus (and also Free) accounts ("bring back 4o, o3, ...")
- Context window lengths aren't clear, nor how many and which exact GPT-5 models ChatGPT uses (standard, mini, new, old) - no indicator of which model is active at any moment
- The gradual rollout and current capacity constraints add to confusion, because not everyone can access the new models yet
- Mixed-up charts during the livestream provided plenty of free material for chart-crime and vibe-charting posts, which hasn't helped build trust or confidence
- And of course, the community had extremely high expectations across all fronts and modalities, as usual
I'm hoping improvements in the next hours/days, and today's AMA, can clarify some of these open issues and questions
Distractosphere retweeted
BREAKING 🚨: High chance that GPT-5 will be released on August 5!
Why?
- August "5"
- GPT-5 related livestreams are being scheduled by YouTubers.
- Google is hosting the Tokyo Next event on August 5 (Another thing to watch for)
Replying to @btibor91
Distractosphere retweeted
🟣New: Qwen3-Coder by @Alibaba_Qwen
- 480B params (35B active)
- Native 256K context length, extrapolates to 1M
- Outperforms Kimi, o3, DeepSeek, and more on SWE-Bench Verified (69.6%) 👀
Now live, starting at $1/M tokens 👇
Distractosphere retweeted
Just realized a new version of Conductor (0.16)!
The terminal panel now persists across workspaces, so you can have multiple long running commands running at once. We also fixed a bunch of bugs.
More coming soon :)
Distractosphere retweeted
An attempt to explain (current) ChatGPT versions.
I still run into many, many people who don't know that:
- o3 is the obvious best thing for important/hard things. It is a reasoning model that is much stronger than 4o and if you are using ChatGPT professionally and not using o3 you're ngmi.
- 4o is different from o4. Yes I know lol. 4o is a good "daily driver" for many easy-medium questions. o4 is only available as mini for now, and is not as good as o3, and I'm not super sure why it's out right now.
Example basic "router" in my own personal use:
- Any simple query (e.g. "what foods are high in fiber"?) => 4o (about ~40% of my use)
- Any hard/important enough query where I am willing to wait a bit (e.g. "help me understand this tax thing...") => o3 (about ~40% of my use)
- I am vibe coding (e.g. "change this code so that...") => 4.1 (about ~10% of my use)
- I want to deeply understand one topic - I want GPT to go off for 10 minutes, look at many, many links and summarize a topic for me. (e.g. "help me understand the rise and fall of Luminar"). => Deep Research (about ~10% of my use). Note that Deep Research is not a model version to be picked from the model picker (!!!), it is a toggle inside the Tools. Under the hood it is based on o3, but I believe is not fully equivalent of just asking o3 the same query, but I am not sure.
All of this is only within the ChatGPT universe of models. In practice my use is more complicated because I like to bounce between all of ChatGPT, Claude, Gemini, Grok and Perplexity depending on the task and out of research interest.
Distractosphere retweeted
Join us from 9:30am PT tomorrow: clau.de/livestream.
Distractosphere retweeted
Replying to @kimmonismus
There was talk about 10 million and 100 million token context lengths, but no real advancements in context have been made. Nothing was mentioned about memory either. The only difference from last year is that large companies are copying the business models of startups. So far, the only somewhat original work, even though it’s technically a copy, just a very well-done one, is @AnthropicAI 's Claude Code.
Distractosphere retweeted
Replying to @512x512
NATURAL CODEBASE MANIFESTO LOADING...
Pure code and natural performance. This software library is handmade by real developers.
Created without AI, built entirely by human hands.
Transparent, sustainable, and made with all-natural code by people, not machines.
Inspired by nature, open-source, and free from any artificial or automated generation.
Just what you need. No unnecessary lines. Pure, organic, hand-crafted code.
Faster and safer. No third-party bloat. No AI shortcuts.
For healthy projects: additive-free, human-optimized software.
Naturally readable code, written by humans. Simple, straightforward, and honest.
No chemicals, no bloat, and absolutely no AI. This is pure, hand-made software.
For the health of your codebase: 100% organic functions, crafted by real developers, not by vibe coding.
Distractosphere retweeted
Replying to @sughanthans1 @Neuralithic
Is there any model that can compete with OpenAI’s Claude 3.7 or 3.5? The existing ones (like o3 or o1) feel incredibly sluggish. GPT-4.1 isn't better than Claude 3.5. Codex isn’t even in the same league as Claude Code. Windsurf SWE-1 benchmarks aren’t anywhere near Claude’s level either. It feels like OpenAI is going to release something, but in the end, no one’s going to be happy with it.
xitter.gq/sama/status/1923104360…
Distractosphere retweeted
Introducing MCPs in Chorus!
Run any server you want with Claude / Gemini / o3.
No need to press confirm when the AI uses a tool.
No need to edit JSON config files.
Distractosphere retweeted
Elon Musk did not mislead about Grok 3.5's release. The early beta was launched for SuperGrok subscribers during May 5-9, 2025, as announced on April 29. Some users expected a public release, causing confusion, but the promise was specific to subscribers. Broader access is planned for June or July 2025.
Distractosphere retweeted
ManusAI was the surprise of the week. It will result in another DeepSeek moment.
It’s pivotal for OpenAI, Anthropic and xAI to come up fast with new developments and stay ahead.
Distractosphere retweeted
OpenAI o3-mini is a good model, but DeepSeek r1 is similar performance, still cheaper, and reveals its reasoning.
Better models will come (can't wait for o3pro), but the "DeepSeek moment" is real. I think it will still be remembered 5 years from now as a pivotal event in tech history, due in-part to the geopolitical implications but for many other reasons too.
All this discussed in 5 hour technical podcast I just recorded on the state of AI industry. Out tomorrow (hopefully).
Distractosphere retweeted
Replying to @AllBiteNoBark88
It is time to use concrete.
en.wikipedia.org/wiki/Portla…
There are too many wooden structures in residential areas in the USA. If they used concrete instead, such situations wouldn't be this severe. Building a structure's framework with wood and using the latest expensive technology for everything else may enhance comfort. Still, regarding safety, a simple fire can destroy massive areas.
Using concrete would be essential in preventing such extensive destruction during fires.



























