Updated
Updated
Showing low-relevance items too. Hide low-relevance items
Thomas Wolf@Thom_WolfAI score38AIOpenAI is releasing a broad range of new mathematical results produced by an internal frontier model, with release guidance from the Institute for Advanced Study's Advisory Group on Mathematics and Artificial Intelligence. The results are available on GitHub at openai/math. The post itself is brief and emphasizes the results rather than hype.
Ethan Mollick@emollickAI score7AIEthan Mollick sarcastically says AI commentators routinely claim deep number theory expertise when opining on the latest math breakthroughs. The post offers no specific breakthrough, model, or figure, so its point is a skeptical observation about AI-community commentary.
AISimon Willison released llm-openai-decisions 0.1a0, a plugin that adds OpenAI's new Decisions API to the LLM command-line tool. The plugin supports yes/no, choices, and score question types, and works with the gpt-6-luna decision model, which accepts both text and image input. OpenAI charges 10 cents per million input tokens for gpt-6-luna, while Jev's rate is 4.2 cents per million, and output is not charged.
Thomas Wolf@Thom_WolfAI score22AIPollen Robotics' Microduck now runs on its first fully custom single-board computer, built by Seeed around the same Rockchip CPU. The new board adds better memory, WiFi/BT, dual NFC antennas, status LEDs, an RGB flashlight, and improved cooling and boot speed, replacing the earlier Radxa-based prototype stack.
roon@tszzlAI score10AIRoon notes that in 2023 GPT-4 was confused by elementary school story problems, a limitation younger users may not remember. The post offers a brief reminder of how quickly model capabilities have advanced since then.
Joshua Achiam@jachiam0AI score14AIJoshua Achiam recommends a deep, carefully considered piece on some of the thorniest problems of our time, regardless of whether readers agree with its prescription. The quoted post, from @satpugnet, presents Phase Lock, a six-month manifesto on how brain-computer interfaces could help align AI and preserve human agency.
Guillermo Rauch@rauchgAI score29AIVercel's AI Gateway now reruns an agent's decision on a fallback model when the primary model's confidence falls below a user-set threshold, now in beta. Rauch calls the simple feature highly impactful for at-scale AI decision-making.
AIPyTorch's blog describes a Triton-based implementation of Table Batched Embedding (TBE) forward and backward kernels for recommendation-system embedding lookups, which the post says outperforms legacy CUDA kernels on these workloads. On B200, an updated CUDA bounds-check step reaches up to 1.24x speedup on that component, and an optional forward-side preprocessing path cuts combined latency from 79.537 ms to 66.183 ms (−16.8%) on a large configuration.
will depue@willdepueAI score62AIA post by Will DePue titled "Fable 5.1's list" presents 100 mathematical results and says 59% were released today, 87% AI and 13% human. The list includes items attributed to OpenAI, Anthropic, Google DeepMind and human mathematicians, each marked by a colored indicator, and it describes many entries as formalized in Lean or as openai/math family numbers. The post supplies no independent verification of these claims.
will depue@willdepueAI score7AIWill Depue, an OpenAI account, asked GPT 6 Pro and Fable 5.1 to rank all discoveries from the last three years. He color-coded them by origin: human, AI before October 6, and AI from OpenAI's math repo. He said 81% of the listed discoveries were released today.
MiniMax (official)@MiniMax_AIAI score7AIMiniMax highlighted its longtime creative partner and client SILVERSIDE in a customer spotlight post. The post credits SILVERSIDE with using MiniMax Music and H3 for animation and lip sync work.
Vercel Developers@vercel_devAI score22Vercel Developers@vercel_devAI score40AIVercel says its AI Gateway can rerun an agent's decision on a fallback model when the primary model's confidence falls below a user-set threshold. The feature is available in beta.
Tibo@thsottiauxAI score46AIOpenAI says its models solved several important math problems and that the team documented the work in a blog post. The post calls this a remarkable new era of scientific progress, but it does not name the specific problems here.
Alex Heath@alexeheathAI score42AIReflection CEO Misha Laskin argues that closed AI models are like renting an apartment, while open models let users own intelligence as AI adoption grows. He says the only way to own intelligence is if it is open. Reflection is preparing to release Beam, its first open-weight model, in a podcast discussion with its co-founders.
Andrew Curran@AndrewCurran_AI score17AIOpenAI has published a new Math post with a linked repository at The post itself gives no further details in the provided text, so the scope of the work cannot be confirmed from this source.
indigo@indigoxAI score29AIindigo (@indigox) says Claude now works directly inside Google Docs, Sheets, and Slides, a feature they had been waiting for. They contrast this with Gemini's integration in Google's own apps, which they call a waste.
OpenAI@OpenAIAI score15OpenAI@OpenAIAI score62AIOpenAI is releasing a broad range of new mathematical results produced by an internal frontier model. The company says it consulted the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study and drew on its advice and public recommendations for how the results are released. The results are available at
lauren@potetoAI score14AILauren Tan notes that Grok Bot, which she often discusses in Slack, can also be added to Microsoft Teams. The post links to an x.ai bot plugin page for the integration.
Teknium 🪽@TekniumAI score33AITeknium announced Hermes Index, which combines scores from the new HermesBench and three other agent benchmarks. The index aims to help Hermes Agent users find the best model at a given time and at a given price point. It was introduced by Nous Research as a way to inform model choice and show labs their performance in Hermes.
AIThe source provides only a title, a source label, and a comments link, with no body text describing the AI mathematics progress. Specific models, results, and benchmarks cannot be verified from the material provided.
Nous Research@NousResearchAI score34AIClaude Opus 5.5 leads the benchmark with a score of 63.31 at $4.99 per task, ahead of GPT 6 Astra at 56.25 ($11.61) and Sonnet 5.5 at 53.14 ($2.82). At the low end, DeepSeek V4.1 Flash scores 36.91 at $0.26, and Ling 3.0 Flash scores 21.56 at $0.054.
Nous Research@NousResearchAI score30AINous Research's Hermes Bench runs every model through the same harness, with reasoning set high where offered. The index reports each model's mean score and mean cost per task across four suites.
Nous Research@NousResearchAI score22AINous Research says its evaluation uses an index of four benchmarks, including its newly introduced Hermes Bench. The other three are Terminal-Bench 4.0, Terminal-Bench-Science, and SkillsBench.
Nous Research@NousResearchAI score31AINous Research has introduced Hermes Index, an average of benchmarks that measures model performance within Hermes Agent. The index aims to help users choose between models and show AI labs how their models perform in Hermes.
Replit ⠕@ReplitAI score7AIReplit is hosting a San Francisco Tech Week discussion tomorrow with Passion Fruit, ElevenLabs, and Gamma on the creator economy. The panel will cover how AI and tech have changed creator-brand partnerships and what true influence looks like. RSVP is available via a Partiful link.
AIPenguin Mail 1.0.5 is a free, GPL-3.0-or-later Rust email and calendar client for x86_64 Linux that handles Gmail, Microsoft, IMAP and POP3 accounts in one inbox. It uses GnuPG for OpenPGP and S/MIME, and its assistant is off until a user picks a model, which can run locally through LM Studio or Ollama.
OpenCode@opencodeAI score20AIMistral Large 4 is now available in OpenCode. The post says there is a 50% discount for the next two weeks.
Vaibhav (VB) Srivastav@reach_vbAI score46AIOpenAI has released its Decisions API in public beta, using GPT-6 Luna to classify text and images, route requests, and score inputs. It is reported to run about 10× faster than the Responses API, starting at $0.10 per million input tokens with no output or cache charges.
GitHub@githubPickAI score72AIGitHub reports that Git events on the platform rose from 218.2 billion to 473.3 billion per month between September 2025 and August 2026. It says agent workloads push write throughput and merge contention beyond what its current replica-based architecture handles well, so it is separating durable storage from compute while GitHub keeps running. The article states internal benchmarks reached up to 35 times higher write throughput.
Why it matters: The post links rising Git event volume to specific architectural bottlenecks, showing why agent workloads strain write paths and how GitHub plans to separate storage from compute.
Ado@adocompleteAI score62AIand those files can also open inside Claude. In Google Workspace, Claude appears in a sidebar next to the open file, reads its contents, and edits it in place, with the option to approve each edit before it is applied.
Google Antigravity@antigravityAI score18AIGoogle Antigravity is promoting its platform, urging users to start building in Antigravity now via a link to antigravity.google. The post gives no further details about features, pricing, or availability.
Google Antigravity@antigravityAI score36AIGoogle's Antigravity agent can take an Android app from prompt to a real device, using the Stitch MCP and Android CLI plugin. The agent pulls designs, builds native Jetpack Compose components, verifies them in the emulator, and runs the final build on a physical phone.
PrismaX@PrismaXaiAI score49AIAt IROS 2026 in Pittsburgh, at least 17 dexterous hand companies exhibited, 11 of them Chinese, with WUJI reportedly shipping 800 to 900 units a month. Boston Dynamics skipped a booth but released a video on the final day of a new four-finger, 13-degree-of-freedom Atlas hand, down from 7 DOF on its previous gripper. Hand makers are also selling capture gloves and data services, since labs need far more demonstrations than the hardware alone provides.
ollama@ollamaAI score18AIOllama posted a link to a model page for embeddinggemma-2 in its library, tagging GoogleDeepMind. The post itself gives no further details on the model's parameters, benchmarks, or capabilities.
ollama@ollamaAI score55AIOllama announced that Google DeepMind's EmbeddingGemma 2 is now available on Ollama. The author describes it as made for consumer devices and multimodal, and gives the command ollama pull embeddinggemma-2 to download it. The quoted DeepMind post says the model is a natively multimodal open model for on-device embeddings that unifies code, images, audio, and video in a shared space.
fal@falAI score13AIfal announced that Google's Nano Banana 2.1 is available today on its platform for both text-to-image generation and image editing. The post links separate pages for each mode but gives no benchmark, pricing, or capability details.
fal@falAI score35AIGoogle's newest image model, Nano Banana 2.1, is now available on fal. The post says it brings faster generation, better subject consistency, improved visual design, and mask-based editing.
Fireworks@FireworksAI_HQAI score28AIFireworks' Head of AI Developer Education fine-tuned a 9B Jev-style classifier on Fireworks using plain SFT, public datasets, and two simple tricks. The full recipe is open source so developers can train their own decision classifiers without a frontier lab budget.