Updated
Updated
Items with an AI score under 20 are hidden. Show low-relevance items
Harrison Chase@hwchase17XAI score28AILangChain's ModelRouterMiddleware uses Jev to read the first message and select a model that handles the entire run. Because Jev is cheap, developers can also re-select a model after each tool result using a custom hook. The router was demonstrated in a quick project by @dbreunig built with DSPy and Jev.
AIIndexTeam released Index-Homura-9B-FP4, an official NVFP4 (W4A4) quantization of the Index-Homura-9B translation model from the Index-Translate family. On a fixed corpus, perplexity rose from 2.5386 in BF16 to 2.6245, a 3.38% increase, and zh->en generations matched the original. Full FP4 compute acceleration requires an NVIDIA Blackwell GPU, while older GPUs get only weight-only memory savings and the FP8 build is recommended for them.
Cohere@cohereOfficialAI score55AIAleph Alpha has released Kolibri, a model with 78B total parameters, 3.46B active, and up to 1M tokens of context, under Apache 2.0. The quoted post says the weights can be run on the user's own hardware. Cohere's own post is a short endorsement of the release.
Aidan Gomez@aidangomezXAI score46AIAlephAlpha has released Kolibri, a German-English model with 78B total parameters, 3.46B active, and context up to 1M tokens. The weights are available under Apache 2.0 for running on users' own hardware. Cohere's Aidan Gomez congratulated the team on the model and its detailed technical report.
Khazix@Khazix0918XAI score38AIThe creator Khazix open-sourced AIHOT, a monthly-active-million AI hot-news site, including its collection workflow, curation scoring, clustering mechanism, and production prompts. The main post reports the project gained 5.1K GitHub stars within a few days and thanks supporters on X.
AIEmergn research claims as many as one in four (23%) UK senior leaders may be bluffing about their AI knowledge. The study says 38% believe their career prospects could suffer unless they significantly improve their AI skills over the next year. Only 43% say they have received substantial AI training in the past 12 months, and Emergn calls for more relevant, human-centric training.
OpenClaw🦞@openclawOfficialAI score23AIOpenClaw v2026.9.8 is released with support for GPT-6.1 Sol, agent replies that return to their original conversation, and lower memory use. The release also includes update and Windows startup fixes, covering 43 PRs from 8 contributors.
vLLM@vllm_projectOfficialAI score31AIvLLM Semantic Router has released Decision 2.0, featuring open models ranging from 0.6B to 27B parameters. The models can answer 64 questions about a request in a single forward pass, are licensed under Apache-2.0, and load with Transformers.
Jerry Liu@jerryjliu0XAI score34AILlamaIndex introduced Extract v2.5, a set of document extraction agents that can reconstruct records split across pages and assemble them with thousands of other cells into structured tabular output. The post says the agents handle real-world documents like insurance claims, regulatory filings, and legal schedules, where a record may start on one page and finish on the next. The accompanying background post claims record-spanning-page accuracy rose from 85.5% to 96.5%, and that the agentic tier outperforms Opus 5.5 and GPT-6 Sol at 30% to 4x lower cost.
OpenClaw🦞@openclawOfficialAI score36AIOpenClaw has integrated Tencent Hunyuan's AI-Infra-Guard (AIG) into ClawScan, its open-source command-line security tool. Every skill and plugin uploaded to ClawHub now runs through AIG as part of its security review.
Stability AI@StabilityAIOfficialAI score22AIStability AI's Executive Chairman Sean Parker and CEO Prem Akkaraju spoke with The Information about the company's work building tools for music professionals. The interview is linked to a report on Stability's push, backed by major record labels, to win over musicians on AI technology.
Pika@pika_labsOfficialAI score31AIPika has redesigned its video creation experience, offering Video Studio for creators who want to focus on craft and direct generation through specific models for those with model preferences. A creator notes that Seedance on the platform costs about half of what they pay elsewhere, with no locked top-tier model behind a higher plan.
Guillermo Rauch@rauchgXAI score34AIMuse has released Muse Gadgets, an open-source ESP32 firmware and Linux SDK for building hardware devices that work with Muse. Developers can obtain an API token from gadgets.muse.ai and use a coding agent with the GitHub repo to build peripherals. Guillermo Rauch praised the team's rapid shipping.
Muse@MuseOfficialAI score33AIMuse has released Muse Gadgets, open-source ESP32 firmware and a Linux SDK for building hardware devices that work with Muse. Developers can get an API token at gadgets.muse.ai and use a coding agent with the GitHub repository to build custom peripherals.
SGLang@sgl_projectOfficialAI score38AISGLang has released v0.5.20, bringing Intel XPU into standard releases alongside RL sampling masks that make rollouts more reliable with up to 52% faster decode. The update also adds Unified Radix Tree SWA branching-point caching, which the project says lifts cache hit rate about 20 points and cuts TTFT by roughly one-third, plus up to 12.5× faster ROCm model loading. New models named in the release include GLM-5.3-Flash, Qwen3.8-Flash-Next, K2 Horizon, Hy4-Preview, FastH3, and VDN-H3.
SGLang@sgl_projectOfficialAI score39AISGLang's update adds a /v1/score endpoint that returns scores for requested labels such as Yes/No or A/B/C, avoiding the label loss of generate with top-k logprobs. Its multi-item scoring computes shared context once and keeps each candidate isolated, with 16-candidate p95 on Qwen3-8B dropping from 54.1 ms (Generate) to 20.6 ms.
SGLang@sgl_projectOfficialAI score28AISGLang demonstrated Qwen3.8-27B as a multimodal decision model that beat Pokémon FireRed's Elite Four and champion with sub-100 ms decisions from live game state. The company says its native /v1/decisions API lets LLMs and VLMs be used for classification and scoring. It also announced /v1/systemone for running Jev-like open models with the TypeSafe SDK.
SGLang@sgl_projectOfficialAI score58AISGLang has released v0.5.21 with a native Decisions API that turns an LLM or VLM into a low-latency classifier and scorer. The release also lets /v1/score rerank search or RAG results in one call, lets PD instances switch between prefill and decode without restarting, and adds support for models including DeepSeek-V4.1 Flash, Kimi K3, and GLM-5.3-Flash on AMD MI355X. The announcement reports a 22% faster first token on long prompts for DeepSeek-V4.1 Flash and 20.6% higher prefill throughput for Kimi K3 in PD serving.
LiveKit@livekitOfficialAI score23AIAssemblyAI's Universal 3.6 Pro speech-to-text model is now available in LiveKit Inference, with 45% fewer wrong yes/no confirmations and about 30% less background speech transcribed. It supports 32 languages plus code-switching and endpointing that waits out phone numbers and emails, at the same $0.45/hr price, accessible by switching to universal-3-6-pro.
Sara Hooker@sarahookrXAI score26AIAdaption Labs has made Invent, its tool for generating AI training datasets from a plain-language description, available through an API. Developers can reportedly produce AI-ready training datasets in minutes with a few lines of code, according to the quoted post. Documentation is available at docs.adaptionlabs.ai.
Cline@clineOfficialAI score38AICline Desktop now offers Connectors in beta, letting users link Gmail, Slack, Google Calendar, Linear, Sentry, Notion, and other apps in one click. Once connected, Cline can use these apps' tools to retrieve context and take actions on the user's behalf.
François Chollet@fcholletXAI score28AIKeras is moving to a pluggable backend design, with MLX and PaddlePaddle backends upcoming as add-on libraries. The team is reducing the operations needed to ship new backends and streamlining unit testing so a single harness can test all ops, such as casting consistency. KerasHub also gains many new models and is shifting its preprocessing from tf-text to PyGrain.
Nathan Lambert@natolambertXAI score35AINathan Lambert and Tom Zick have unveiled Trillium Labs, a new non-profit focused on the open science of frontier AI. The lab plans to build open post-training recipes and expand into open infrastructure to study topics such as RSI, reward hacking, and multi-agent systems. It is hiring, fundraising, and seeking compute, with support from Halcyon Futures and Schmidt Sciences.
Jerry Liu@jerryjliu0XAI score44AILlamaIndex released Extract v2.5, a set of document extraction agents that it says reach 93%–96%+ accuracy on long-list extraction, including records spanning pages. The post claims the agents outperform frontier VLMs, which it says stop early, miss repeated records, and struggle to attribute values to sources, while LlamaIndex attributes every extracted value to its source. The agents are available through LlamaParse.
ComfyUI@ComfyUIOfficialAI score34AIComfyUI released YUI, described as the first animated short film made with Comfy Agent, a 4:20 film its creator made in three days. Comfy Agent handled shot regeneration, side-by-side model comparisons, and character consistency checks that previously required weeks of manual testing and generation.
Krea@krea_aiOfficialAI score38AIKrea has made FLUX 3 Image available on its platform. The model supports multi-turn edits that leave all other pixels unchanged, layout via bounding boxes, generation up to 4K, and composition with up to 10 references.
Vercel Developers@vercel_devOfficialAI score24AIVercel says the Laya decision model is available free on AI Gateway through October 31, in partnership with Boundless. The post suggests using it to route agent work, triage support requests, and check guardrails.
Fidji Simo@fidjissimoXAI score20AIFidji Simo says she is excited to work with Morgan Cheatham and Jimi Hendrix, two early investors who recognized the importance of data in AI biotech. Morgan Cheatham's linked post says the partnership supports ChronicleBio, which is building AI models to identify biologically distinct patient subgroups within complex chronic diseases. The team reportedly combines clinical phenotyping, molecular measurement, and AI, and its early work has surfaced disease subtypes linked to potentially relevant therapies.
Fidji Simo@fidjissimoXAI score22AIFidji Simo's company ChronicleBio aims to develop treatments for POTS, Long COVID, ME/CFS, and hEDS, which currently have zero FDA-approved therapies. The company plans to build the deepest dataset on these conditions and use AI to identify the biology behind their symptoms. The goal is to match the right drugs to the right patients.
Pika@pika_labsOfficialAI score23AIPika has released a set of new creation apps on its updated platform at pika.art. The post itself gives no further details about the apps' features or capabilities.
Pika@pika_labsOfficialAI score34AIPika has launched Camera Director, an app that reframes any shot to show it from different camera angles while preserving the original content. A quoted post praises its quality and says it is useful for real-world work.
Pika@pika_labsOfficialAI score31AIPika has launched a new AI creative platform offering more than 40 apps covering needs from finishing edits to full video ads. The company describes the product as built for creatives and designed to meet exacting output standards.
Replicate@replicateOfficialAI score46AIReplicate says it is powering Griffin from Tavus on its platform. Tavus describes Griffin as the first model to pass the video Turing test, with 48% of live conversation participants believing it was a real human. The model ranks first on NVIDIA's full-duplex AI video benchmark.
AIApodex has released Apodex 1.1 Mini, a free reasoning-first model designed for complex, long-horizon research and forecasting tasks. According to the source, it works directly with files, data, code, and tools to produce verifiable results.
ComfyUI@ComfyUIOfficialAI score24AIComfyUI announces the Comfy Developer Platform, which includes the Comfy API and Comfy Router for building applications. The post is a broadcast link without further details on features, pricing, or availability.
Prime Intellect@PrimeIntellectOfficialAI score32AIExtropic post-trained Qwen3.6-35B-A3B with Prime Intellect for thermodynamic ML research, nearly tripling its held-out eval results in about 100 GRPO steps. The team built a custom RL environment with verifiers and trained on Hosted Training, Prime Sandboxes, and Prime Inference. This let Extropic avoid managing multi-node GPU infrastructure and focus on research.
AIBenchmark has won a hot competition to lead a new funding round for early-stage chip startup Tendrils Compute, according to multiple people familiar with the matter. Sources say the company is already discussing a fast follow-up raise that could value it at more than $1 billion. The deal is part of a wave of VC investment in specialized chips, including inference-focused startups such as Etched, which doubled its valuation to $21 billion in August.
ComfyUI@ComfyUIOfficialAI score20AIComfyUI announces Comfy Agent, which its post presents as the first agent built for creative work. The post directs readers to a blog for more details, but its own text provides no further specifics on capabilities, availability, or pricing.
Praan Inc@praan_incXAI score30AIPraan launched a new HIVE, its medical-grade indoor air purifier, priced at ₹49,999 (all-inclusive, $520). The new model is 62mm smaller, more capable, and autonomous, and is 30% more affordable than its predecessor. The company says the HIVE is now in more than 1,500 locations across 8 countries.
Jerry Liu@jerryjliu0XAI score42AILlamaIndex introduced Extract v2.5, a series of agents tuned for document extraction across cost-effective, agentic, and agentic plus tiers. The company reports the agents outperform Opus 5.5 and GPT-6 Sol while costing 30% to 4x less, with accuracy gains on long lists (86.1% to 95.5%), multi-page records (85.5% to 96.5%), and scanned forms (90.9% to 95.7%) on its agentic tier. The release adds advanced citations with bounding boxes and structural reasoning, and the agents are available on LlamaParse.
