[NEW MODELS!] Supra2-100M Base and Instruct – go check them out!
Hey guys! After a LOT of good feedback on our previous models like Supra-50M-Instruct and -Reasoning, many community likes, follows and upvotes we saw many community…
Communities 6816 stories
LessWrong vs. TikTok: Tips for capturing attention in a non-rational space
Here's what I've learned from making TikTok videos every day for the last 60 days. I think posting on TikTok is worthwile whether you want to…
ZX Spectrum System Tour: Sound
Article URL: https://bumbershootsoft.wordpress.com/2026/08/01/zx-spectrum-system-tour-sound/ Comments URL: https://news.ycombinator.com/item?id=49159676 Points: 4 # Comments: 0
Qwen3.8-Max matches Kimi K3 and DeepSeek V4 Flash
Qwen3.8-Max (2.4T) is another massive contribution to the open weight community. On benchmarks, it performs closely to Kimi K3 and DeepSeek V4 flash across all categories…
Thermal paste PSA for old GPUs
I know many of us are using older GPUs like the 3090 because they work great. I just replaced the thermal paste and am seeing consistently…
Show HN: Hacker News with AI stories filtered out
Hi folks,I've noticed that a lot of people here seem exhausted by the amount of AI news on the front page. I shared hcker.news here a…
The Billable Usage API: programmatic cost visibility for Cloudflare
Article URL: https://blog.cloudflare.com/billable-usage-api/ Comments URL: https://news.ycombinator.com/item?id=49158714 Points: 13 # Comments: 2
Smaller, faster, safer: running Kimi and GLM at scale
Article URL: https://blog.cloudflare.com/smaller-faster-safer-models/ Comments URL: https://news.ycombinator.com/item?id=49158581 Points: 12 # Comments: 0
Celebrating 45 Years of Kermit with the First New C-Kermit Release in 15 Years
Article URL: https://changelog.complete.org/archives/44456-celebrating-45-years-of-kermit-with-the-first-new-c-kermit-release-in-15-years-and-working-with-a-decades-old-c-codebase Comments URL: https://news.ycombinator.com/item?id=49158474 Points: 27 # Comments: 4
Show HN: Run an 80B Qwen in 4.3 GB of RAM on a Mac, and a 35B on an iPhone
Article URL: https://github.com/leonickson1/Swiftlet Comments URL: https://news.ycombinator.com/item?id=49158333 Points: 4 # Comments: 0
Use Task Runners for Common Coding Tasks
Article URL: https://hamvocke.com/blog/task-runners/ Comments URL: https://news.ycombinator.com/item?id=49158287 Points: 13 # Comments: 1
Help Wanted
Article URL: https://lake.computer/blog/help-wanted/ Comments URL: https://news.ycombinator.com/item?id=49158214 Points: 8 # Comments: 0
The Chinese labs everyone lumps together are making four pretty different bets. I work at one of them.
Every time a model drops from a Chinese lab the thread fills with people who already know who made it, and the guess is usually Alibaba.…
SearXNG in Rust
Article URL: https://github.com/MikeLuu99/searxng-rust Comments URL: https://news.ycombinator.com/item?id=49158141 Points: 12 # Comments: 2
Launch HN: Hoplite (YC S26) – Effortlessly deploy cloud coding agents
Hi HN, we’re Bence and Ryan, founders of Hoplite (https://hoplite.sh). Hoplite lets you deploy coding agents in the cloud, with a suite of tools that makes…
Show HN: Product analytics (and evals) for agent sessions on your MCP
Hi HN! We’re Theodore and Louis, founders of Armature (YC P26). We reconstruct the entire session behind the MCP tool calls you receive, including what the…
It’s time to desk reject papers that don’t include code that can reproduce the results [D]
As review season for NeurIPS wraps up, I have now reviewed for 3 major conferences this year. And I'm noticing a worrying trend: Out of the…
What's the largest software project AI can complete on its own?
Article URL: https://epoch.ai/MirrorCode Comments URL: https://news.ycombinator.com/item?id=49157786 Points: 15 # Comments: 4
Explanation of INT8 ConvRot (FP8 is no longer needed)
Article URL: https://note.com/hirorohi03/n/n047a8c5f7f8b?hl=en Comments URL: https://news.ycombinator.com/item?id=49157645 Points: 6 # Comments: 2
I CANNOT believe I’ve got DeepSeek-V4-Flash-0731, a frontier model, running on my home PC. Insane!
So this is the stuff of absolute insanity. In less than 20 months we've gone from super expensive cloud models only, to being able to run…
an espresso Q/A model running fully offline on an ESP32S3
i already had an esp32 generating stories, but generating text is not the same as receiving a question and giving a useful answer. barista v0.1, a…
Taylor Farms Has Rewritten Its Cyclospora Statement Four Times in Sixteen Days
Article URL: https://www.marlerblog.com/case-news/taylor-farms-has-rewritten-its-cyclospora-statement-four-times-in-sixteen-days-it-still-has-not-said-what-changed-at-that-plant-after-2013-or-why-two-thousand-negative-tests-should-mean-an/ Comments URL: https://news.ycombinator.com/item?id=49157128 Points: 69 # Comment
All DeepSeek model oneshots: 242 outputs to look at and compare!
Continuing my weekend of oneshotting the cheap OpenRouter models, here are all 10 DeepSeek models across the same 35 prompts. DeepSeek had a rougher time (more…
DeepSeek V4 Flash 0731 – Happy Numbers (700pp/18tg) and Thoughts
Originally, I was only getting around 140pp/s and about 21tg/s, but the config with -b 8192 -ub 8192 --cpu-moe is vastly superior, let's say 700pp/s and…