best vlm or llm for 192gb ddr5 (dual channel) + 2x 5090?
hey, haven't been up to date for longer about what currently are the best models. so whats the best recommended model for this? can be either…
hey, haven't been up to date for longer about what currently are the best models. so whats the best recommended model for this? can be either…
Article URL: https://www.jamesdrandall.com/posts/thrust_ai_powered_software_archaeology/ Comments URL: https://news.ycombinator.com/item?id=48865903 Points: 5 # Comments: 0
Hello all, currently have my strix halo with windows and 96gb vram memory reserved. Which I thought was the proper way to do it until I…
Apple has sued OpenAI, alleging that former employees that now work for the AI company have stolen Apple's trade secrets "for the benefit of OpenAI." In…
I've been going on about how ChatGPT Work (and Cowork) are missed opportunities for knowledge workers, and to illustrate that take a look at Google's NotebookLM…
RT Andrew MilichBuilt with two prompts and /goal!Jayden Davis: Grok 4.5 in Grok Build is insane for 3D game dev🤯 I ran a detailed prompt using…
(The aspirant sequence (so far: Would You Work Harder In The Least Convenient Possible World?, That Which Cannot Be Poked With A Stick Is The Mind-Killer)…
I used to check LM Arena anytime new open models came out to see how they stacked up to the big closed ones. But it seems…
Article URL: https://www.nytimes.com/2026/07/10/technology/apple-openai-lawsuit.html Comments URL: https://news.ycombinator.com/item?id=48865431 Points: 65 # Comments: 10
Article URL: https://www.ycombinator.com/companies/moss/jobs/52LnqLQ-software-engineer-sdk Comments URL: https://news.ycombinator.com/item?id=48865332 Points: 0 # Comments: 0
Try Grok 4.5 in PerplexityPerplexity: Grok 4.5 is now available as an orchestrator model in Computer for Consumer Pro and Max subscribers.We evaluated it against five…
Article URL: https://arxiv.org/abs/2607.08147 Comments URL: https://news.ycombinator.com/item?id=48865238 Points: 3 # Comments: 0
Apple alleges the misconduct was directed by OpenAi's senior leadership, including a long-time former employee.
GPT-5.6 is a major step forward for health intelligence.Across the lineup, we’re delivering stronger performance at lower cost: GPT-5.6 Luna outperforms GPT-5.5 at its highest reasoning…
Token usage beginning to tip towards open models. Ollama’s @jmorgan and Peter Fenton predict that a supermajority of tokens in the future will be from open…
Article URL: https://www.tryai.dev/blog/gpt-5.6-build-off-12-models Comments URL: https://news.ycombinator.com/item?id=48865093 Points: 55 # Comments: 27
https://www.macrumors.com/2026/07/10/apple-sues-openai/ Comments URL: https://news.ycombinator.com/item?id=48865019 Points: 74 # Comments: 34
Article URL: https://nebusec.ai/research/ionstack-part-2/ Comments URL: https://news.ycombinator.com/item?id=48864969 Points: 5 # Comments: 1
i made a multiple linear regression trainer that can be used with custom data in scratch nothing more to say, the impressive part is the scratch…
Article URL: https://www.nytimes.com/2026/07/10/business/atlassian-tech-worker-wrongful-termination.html Comments URL: https://news.ycombinator.com/item?id=48864825 Points: 22 # Comments: 8
MuScriptor is an open-weight, decoder-only Transformer from Kyutai and Mirelo. Trained on 170k real recordings plus 1.45M synthetic MIDIs, it transcribes full multi-instrument mixes into MIDI.…
server: improve tools, remove apply_diff (#25498) server: improve tools, remove apply_diff improve edit tool add tools_io abstraction add tools_io_basic fix build move utils to class member…
server: accept null sampling params Extend the schema validation to treat a null value as absent, so clients can send null on nullable params (temperature, top_p,…
And then what is the point of Work? Just a dumbed-down version of Codex that secretly does coding but hides it?(I can understand, maybe, if the…