zach

Father x2. Software Engineer and AI operations consultant building solutions for small and large enterprises alike.

[zach @epidev]

Love seeing content like this. These models can run on hardware most people can actually get

[Mia @MiaAI_lab]

Orinth-1.0-35b vs Qwen-3.6-27b

I've asked both to build a single html file of Tetris game. First here's the Orinth 35b

I think it did a fine job. Every control works, the game looks fine visually.

How did Qwen 3.6 27b did? See post #2

[Google Gemma @googlegemma]

Gemma 4 is the first multimodal model on Cerebras! ️

What can you build with Gemma 4 31B running at 1500 tokens per second?

Join the Cerebras x Gemma 4 24-hour virtual hackathon this Sunday to compete for $5,000 in prizes.

Participants get early access to Gemma 4 on Cerebras.

[zach @epidev]

Interesting that the CLI which only works on open models isn't open itself.

[Tom Greenwald @tomgreenwald]

Magnitude scores 75.5% on Terminal-Bench 2.1, making it the top coding agent for GLM 5.2.

npm i -g @magnitudedev/cli

See how we did it (and kept it fair) below ⬇️

[zach @epidev]

@patience_cave @astralmatrix nothing got released, it's okay! *weeks* they said. patience is still at hand.

[Ivan Fioravanti ᯅ @ivanfioravanti]

ROYAL RUMBLE: M3 Ultra vs M5 Max vs Spark ⚡️ 🧵
- Model: deepreinforce-ai/Ornith-1.0-35B-GGUF:Q8_0
- Server: llama.cpp (9824) - args used i comments
- Cache not used to simulate a single large context load each time

Notes:
- On large contexts DGX Spark on llama.cpp is strong, but again keep in mind that it's not so common starting with 128K context in first message, especially in coding sessions
- In the detailed charts below for DGX Spark you see M3 Ultra because I ran the benchmark remotely, working to fix this.
- I will test with MLX soon using 8bit 💪

Enjoy results and share your feedback or requests for more tests!

[zach @epidev]

@daradoescode Dream big!

[zach @epidev]

@Madisonkanna She's right about it being the hardest and best thing (and I'm just the man)

[zach @epidev]

@0xSero Exciting stuff!

[Vaibhav (VB) Srivastav @reach_vb]

True story: I stopped thinking about context since GPT 5.3 Codex

Single project focused threads with the recent capability of codex to spinoff new threads is goated!

Codex continues and goes through compaction but remembers all the important stuff and if not, it’ll look up through the session and find the relevant info

This is also the reason why /goal is so effective as well

[steve2bacon, CMT 🥓🍩 @steve2bacon]

the united states as soon as the stock market is closed for the weekend

[zach @epidev]

for 20 companies.

[Adam.GPT @TheRealAdamG]

Valuemaxxing

[Brad Gerstner @altcap]

Fable is currently export controlled & rumors are that 5.6 will also be subject to an approval framework. Whatever jiu jitsu the Chinese are using to get us slow down our own frontier models while letting their models run free appears to be working. Who is capturing who? 🧐🇺🇸

[zach @epidev]

strategy change: @sama @OpenAIDevs - hire me :)

[the tiny corp @__tinygrad__]

The US AI pay-to-play scam is so much more tolerable after switching to a locally hosted GLM-5.2. From the front page of HN, open weights will be the frontier this December. Sorry about your IPOs.

[Derya Unutmaz, MD @DeryaTR_]

I am so deeply disappointed by the US gov. It was apparently a mistake to think this administration had a good sense to support American AI models.
However, I also put more blame on the cult of Anthropic & Dario Amodei. All that fearmongering both achieved its goal & backfired!

[BridgeMind @bridgemindai]

Open source just passed American labs in market share.

US models on OpenRouter collapsed from 73% to 33% in one year.

Chinese open source models surged.

Then the US banned Fable 5.

Then gated GPT 5.6 customer by customer.

We are watching the US hand the developer market to China in real time.

[AshutoshShrivastava @ai_for_success]

"AI will benefit all of humanity" aged like milk.

[zach @epidev]

everything including UI? interesting.

[Mikhail Parakhin @MParakhin]

Not as relevant now :-(: I had an opportunity to deeply test both Fable 5 and GPT-5.6 Max. 5.6 is clearly better than Opus 4.8 at everything (slightly faster, too, though that depends on the load). Vis-a-vie Fable, it is clearly worse on coding, but better on agentic workloads. I had Fable write code, 5.6 run experiments - dreamy…

[Anthony Ronning @anthonyronning]

@romainhuet OpenAI employees looking around for the "people building on 5.6"