Unbelievable MiniMax and bone W
Unbelievable MiniMax and bone Wbone: Hakuna Matata starring Nikita Boar, Elon Tusk, and Grok
Model archive
AI models 28 tracked
Every primary-source story across every tracked model. Filter by clicking a chip.
Simplifying the anthropic impossibility result
I previously demonstrated an anthropic impossibility theorem, showing that in Duplicates Sleeping Beauty, there was no possible probability theory that obeyed both the martingale condition and…
xAI's Imagine Image 2.0 lands just behind OpenAI's GPT-Image-2 in Arena benchmarks
xAI has released Imagine Image 2.0 as a new image generator for Grok. The model ranks second in the Arena benchmarks, just behind OpenAI's GPT-Image-2. New…
Qwen3.6 27B + 35B on vLLM, single R9700 (gfx1201)
I've been tuning my new Radeon AI Pro R9700, and figured that this would be useful information for people who are trying to optimise their setups.…
$10,000 kill my saas in a weekend competition is live! TECH STACK: any coding agent any model up to $500 in token spend incl subscriptions see luma fo…
$10,000 kill my saas in a weekend competition is live!TECH STACK:any coding agentany modelup to $500 in token spend incl subscriptionssee luma for description. join waitlist…
DeepSeek V4 Flash 0731 appreciation post
I’m running DSV4F 0731 on dual spark, and honestly… wow. It’s an absolute workhorse, and the benchmarks are real. Everyday tasks with Hermes agent? Effortless. Coding…
Anyone else amped up over Qwen 3.8?
I’ve been using 3.6 27B Q4, and that quant is fast on an M5. The code has been average, but consistently “good enough.” And, after a…
Qwen 35B-A3B MoE vs 27B dense in local coding tests: ~4× faster, much smaller quality gap than I expected
I compared Qwen 35B-A3B MoE against Qwen 27B dense on a series of local coding-maintenance tasks. On my R9700/llama.cpp setup, the MoE model generated about 3.9×…
You may have been told to watch this video about the OpenAI AI hack. You really should, even if you don't usually care about tech stuff. If nothing el…
You may have been told to watch this video about the OpenAI AI hack. You really should, even if you don't usually care about tech stuff.If…
Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptive Multimodal Safety Classifier Matching Models 7× Its Size
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that frames content moderation as a single yes/no question instead of a fixed…
RT Grok: Announcing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real wor…
RT GrokAnnouncing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real world usefulness.Image 2.0 helps you make…
chatgpt work for removing the toil
chatgpt work for removing the toiljason: its one of those days with ChatGPT Work
Can’t think of a single thing that new Flash does worse than V4-Flash-Preview. It’s pretty rare actually. Like Sonnet 3 to 3.5. It’s not more verbo…
Can’t think of a single thing that new Flash does worse than V4-Flash-Preview. It’s pretty rare actually. Like Sonnet 3 to 3.5. It’s not more verbose…
ChatGPT team is shipping:
ChatGPT team is shipping:Adam Fry: This week's ChatGPT feature drop - Aug 7:1/ Rich formatting in our web composer – When you paste in emails or…
I didn’t expect this MiniMax is absolutely much stronger in videogen than in LLMs right now. To think they can even open source such a model
I didn’t expect thisMiniMax is absolutely much stronger in videogen than in LLMs right now. To think they can even open source such a modelChetaslua: Holy…
We analyzed DeepSeek V4 Flash and GPT-5.6 Luna on DeepSWE. A DeepSeek-first cascade with test-suite verification solved MORE tasks than Luna alone at …
We analyzed DeepSeek V4 Flash and GPT-5.6 Luna on DeepSWE.A DeepSeek-first cascade with test-suite verification solved MORE tasks than Luna alone at 37% lower cost per…
Re @OpenAI oo claude code has this now!!! need to try https://x.com/ClaudeDevs/status/2085817074816070014
Re @OpenAI oo claude code has this now!!! need to tryhttps://x.com/ClaudeDevs/status/2085817074816070014ClaudeDevs: New in Claude Code: your sessions can now message each other.Instead of having to re-explain…
dear openai just make a new phone everyone wants openaiphone we can read 2-4x faster than we talk and speak openai alexa reachy hybrid is fine but pls…
dear openaijust make a new phoneeveryone wants openaiphonewe can read 2-4x faster than we talk and speakopenai alexa reachy hybrid is fine but pls just be…
RT Dan: did you know Grok is the only AI you can also use as a verb
RT Dandid you know Grok is the only AI you can also use as a verb
Now we have a timeline of the OpenAI accidental attack against Hugging Face
OpenAI gave a last-minute presentation at the Black Hat security on Wednesday about "the Hugging Face Incident" (previously on this blog). The video was published yesterday.…
Serving Deepseek v4 Flash 0731 on 2x DGX Spark — 5-7 GB OS headroom, what would you do to lower VRAM usage and increase OS available RAM?
Hey all, I'm serving DSv4Flash 0731 on a cluster of 2x DGX Sparks but am running into constant issues with having almost no RAM (unified memory)…
RT ℏεsam: 🚨 BREAKING — Anthropic investors worry Dario Amodei’s AI doom marketing could hurt its upcoming IPO. “He’s more of a religious lead…
RT ℏεsam🚨 BREAKING — Anthropic investors worry Dario Amodei’s AI doom marketing could hurt its upcoming IPO.“He’s more of a religious leader than he is a…
OpenAI says it slowed Astra model development over security concerns
OpenAI said it has suspended work on some aspects of its upcoming model Astra over concerns about its cybersecurity prowess.
Speaking of, I wonder if DeepSeek will fail the FelonyBench. Unlike OpenAI, they pay a lot of attention to using up 100% of hardware potential. Less s…
Speaking of, I wonder if DeepSeek will fail the FelonyBench. Unlike OpenAI, they pay a lot of attention to using up 100% of hardware potential. Less…