explainx / blog / topics
Open-Weight Models
Open-weight models publish their weights so anyone can run, fine-tune, and study them. Releases from DeepSeek, Qwen, Zhipu GLM, Moonshot Kimi, Mistral and others often close the gap with closed models within months.
This page tracks each release we covered, with benchmark results and practical notes on running them.
300 stories · latest Nov 14, 2026
Start here
How to Run Kimi K3 Locally — Confirmed Hardware Tiers and vLLM Setup (2026)
Moonshot's Kimi K3 open weights went live July 26, 2026. This guide covers what actually changed once the download landed — confirmed license, real disk and VRAM numbers, a working vLLM serve command instead of a placeholder, and where community GGUF quants stand two days after release.
Small Models Compared: Haiku 5.5 vs GPT-6 Luna vs DeepSeek vs Gemini Flash vs GLM
Claude Haiku 5.5 landed at $0.10 per million input tokens, the same sticker as GPT-6 Luna. Does that make the small-model tier a tie? We compare five cheap models on list price, cache pricing, context, licensing and the benchmarks that can honestly be compared, and say which to use for subagents, coding, long context and self-hosting.
Top 10 Open-Weight Models You Can Actually Run on a Laptop
A model being downloadable does not make it laptop-friendly. This ranked guide starts with memory math, then recommends ten models that remain useful after weights, context cache, and operating-system overhead are counted.
What Are LLM Parameters? Top 10 Model Sizes (July 2026)
Parameters measure how many learned numbers sit in a model checkpoint — not tokens, not context length. explainx.ai explains total vs active MoE counts, why closed frontiers hide size, and ranks the top disclosed LLM sizes as of July 2026, led by Kimi K3 at 2.8 trillion.
Meta Llama 4: The Complete Open-Source AI Model Guide 2026
Meta Llama 4 is the most significant open-weight model release in AI history. Scout fits on a single consumer GPU with a 10 million token context window. Maverick outperforms GPT-4o on key benchmarks. And Behemoth — the 2-trillion-parameter teacher — is rewriting what we thought was possible outside a closed lab. This is your complete guide to the Llama 4 family.
What are parameters in a large language model? Billions, MoE, and what 2026 model cards really say
Bigger is not a synonym for smarter, but parameter count is still the first axis people use to compare scale. This guide explains what parameters are, how mixture-of-experts changes the math, and which flagship models still publish size—and which do not.
Timeline
November 2026
Nov 14
Open-Source AI Without Jensen Huang: What Changes for BuildersAfter Jensen Huang resigned as Nvidia CEO on November 12, 2026, Interconnects editor Nathan Lambert published an essay on who, if anyone, carries the US open-weight mantle. Here is the argument, what explainx.ai has already verified about Huang, and what builders should watch.
Nov 13
ERNIE 5.5 Flash Preview WebDev Rank: Claimed vs VerifiedOn November 11, 2026, Baidu released ERNIE 5.5 Flash Preview atop the ERNIE 5.5 Flash Base MoE. Arena and X posts call it the #2 open model on WebDev; the leaderboard row is Proprietary with AutoEval and N/A votes.
Nov 13
Baseten DeepSeek Ultrafast: Rust Engine Hits 2,060 tok/s on V4.1-FlashOn November 13, 2026 Baseten published deepseek-ultrafast on GitHub — a Rust inference stack tuned for DeepSeek-V4.1-Flash on sm_100a and sm_103a GPUs. The README documents 2,060 tokens per second decode on eight B300s, million-token prefill in under 12 seconds, and multi-server KV cache transfer over RDMA. Here is what shipped and who should try it.
Nov 13
DeepSeek Opens Its App Features as APIs: Search, Files, URLs, Voice and Safety ChecksDeepSeek said on November 13, 2026 that the features behind its consumer app are now available to developers: web search, file input, URL reading, voice and a safety check. Flags on the existing APIs turn them on.
Nov 12
Alice Huang Firms and DeepSeek: What The Information Claims About Blackwell ChipsTechmeme and social feeds amplified a November 12 report that companies tied to Alice Huang invested in DeepSeek with promises of banned Nvidia Blackwell chips. We map the claim to primary export rules and earlier Megaspeed reporting without treating paywalled detail as fact.
Nov 12
Hugging Face 3D LoRA Guide: Fine-Tune TRELLIS.2 on Your MeshesHugging Face released scripts to fine-tune TRELLIS.2 on custom 3D assets via LoRA, targeting shape errors like incorrect shoe soles. The guide walks pipeline stages, what to freeze, and how to run on Hugging Face Jobs or local Docker.
Nov 12
Hugging Face PyBot: A Rule-Based Chatbot Built by 1,000 AgentsHugging Face released PyBot on GitHub November 12, 2026: a chatbot written in roughly 500,000 lines of Python if-else rules, regex and lookup tables, produced by more than 1,000 coding agents across 5,000 iterations. With no model weights, it reaches about 26% win rate on SmolTalkBench, comparable to SmolLM-360M-Instruct in the README claims.
Nov 12
IBM Granite LLMON: Structured Context Markup for Smaller ModelsOn November 12, 2026 IBM published LLMON (LLM Object Notation) with two Granite 4.1 preview checkpoints on Hugging Face. The format marks instructions, data, and requirements with reserved tokens and exec blocks so 3B–8B models ignore poisoned tool output more often. Here is how it differs from JSON, how training worked, and how to test the previews.
Nov 11
Qwen4-Max Preview Lands on Alibaba Token Plan: What Builders Can Try TodayOn November 11, 2026 Alibaba’s Qwen team said Qwen4 is arriving and put Qwen4-Max into preview on Token Plan and four consumer apps. There is still no public weight drop, API list price row, or full four-model rollout—but builders with a Token Plan can call the preview id today.
Nov 11
BotPsy 0.8B: Tether's Open On-Device Agent Model for Tool CallingOn November 11, 2026 Tether AI Research published BotPsy, a 0.8B-parameter agentic language model built on Qwen3.5-0.8B with Apache 2.0 weights for research and education. Here is what the Hugging Face post reports, how it was trained, and how to try it without sending prompts to a cloud API.
Nov 11
ByteDance Frontier AI by 2030: Zhang Yiming's Timeline, Claimed vs VerifiedOn November 11, 2026, Bloomberg and relay outlets report that ByteDance co-founder Zhang Yiming told an internal AI meeting a personal agent like Meta Muse could ship in weeks, but matching OpenAI and Anthropic at the frontier may wait until 2029 or 2030 because Seed pretraining data lags rivals.
Nov 11
China Drugmakers Pivot to Physician AI Apps After Graft CrackdownOn November 11, 2026 Reuters reported that Chinese drugmakers are exploring AI medical assistants from Alibaba Health and JD Health after Beijing tightened hospital visits. CTO Wang Xiangzhi described a Hydrion model like OpenEvidence sponsorship; 93% of surveyed doctors now use AI tools.
Nov 11
Token Pooling-Aware Training: Cheaper ColBERT-Style Search on Hugging FaceMulti-vector retrievers like ColBERT assign one embedding per token, which explodes index size. A November 11 Hugging Face blog trains models to stay accurate after Ward clustering merges similar token vectors, with open checkpoints on the Hub.
Nov 11
HyperFlow H3 Flash: 14B Open Weights for Joint Video and Stereo AudioHyperFlow H3 Flash is Video Rebirth's few-step distillate of MiniMax-H3. The Hugging Face repo ships GGUF quants, INT8 convrot safetensors, and a Python CLI that generates synchronized video and audio in as few as four diffusion steps without classifier-free guidance.
Nov 11
Kimi K3.1 Pro Highspeed on Amazon Bedrock: Model IDs, Caching, Flex Tier, and OpenCodeOn November 10, 2026, Amazon announced Kimi K3.1 Pro Highspeed from Moonshot AI on Amazon Bedrock. The post walks through inference profile IDs, Chat Completions and Converse calls, Flex and explicit prompt caching, data-boundary promises, and the official OpenCode recipe for terminal coding agents.
Nov 11
Sakana AI Hires Ex-NSA and CIA Veterans as Defense AdvisorsOn November 11, 2026, Sakana AI announced that retired U.S. Navy Admiral Michael Rogers, former NSA and Cyber Command chief, and Tom Moran, a former CIA Chief of Station, joined as senior advisors. They will advise on strengthening defense capabilities and cybersecurity as Sakana expands hiring in Tokyo.
Nov 10
Hugging Face Decision Models: Jev-Style Routing Without JSON SpamLarge language models are great at generating paragraphs and terrible at guaranteeing schema-shaped routing decisions. Hugging Face formalized decision models with a shared API across Transformers and llama.cpp so agents can pick billing vs support without decoding JSON.
Nov 10
Hugging Face Hub Billing via Google Cloud MarketplaceHugging Face published a November 10, 2026 guide for paying Hub usage through your existing Google Cloud account. Enterprise admins link an organization, pass marketplace review, and route Jobs, Inference Endpoints, Inference Providers, GPU Spaces, and Team or Enterprise subscriptions onto the GCP invoice. Here is the workflow and the billing caveats.
Nov 10
Hugging Face Warns Cyberattacks Could Make Open AI Hubs Less OpenOn November 10, 2026, TechCrunch quoted Hugging Face chief policy officer Irene Solaiman warning that cyberattacks could push open AI platforms toward identity gates and upload restrictions. The piece ties OpenCode API key theft, this summer's agent breach, and Fireworks' weights incident into one supply-chain story.
Nov 10
Moonshot AI B300 Super Micro Lease: Beijing Filing, What Bloomberg ReportsOn November 10, 2026 Bloomberg reported that documents filed with Beijing regulators show Moonshot AI rented 32 Super Micro Computer servers equipped with Nvidia B300 GPUs in August. The story lands amid export-control scrutiny and the Jensen Huang probe, but Moonshot and regulators have not confirmed the filing on the record.
Nov 10
ZONOS3-Mini: Zyphra's 100M-Active Open TTS for Browser and SGLangZyphra shipped ZONOS3-Mini, a compact MoE text-to-speech model with about 100 million active parameters, Apache 2.0 weights on Hugging Face, and runtimes for WebGPU browsers, MLX, ggml, and SGLang-Omni. The release targets real-time voice cloning on laptops and phones without cloud-only APIs.
Nov 9
StepFun P²-MOPD: Multi-Teacher Distillation With Parameter Priors on Qwen3The P²-MOPD paper on Hugging Face and arXiv adds expert task vectors as learnable parameter priors during multi-teacher on-policy distillation. Authors report faster, more balanced domain recovery on Qwen3 1.7B and 4B models, with code at SinuoWang/P2-MOPD.
Nov 8
DeepSeek Harness Hits 1.54M Daily Users: Creator Mode and a Plugin MarketplaceDeepSeek Harness crossed 1.54 million daily active users on the official API as of November 4, 2026, with about 35,000 community plugins and a new Creator Mode that builds plugins from plain-language requests. The team is planning an official marketplace and creator incentives. Here is what changed, how it compares to other harnesses, and how to try it without treating vendor growth stats as gospel.
Nov 7
How to Read a Hugging Face Model Card (Without Trusting the Headline)Every open-weight release ships a README model card on Hugging Face. The YAML header decides discoverability; the body should document limits, evals, and how to run the checkpoint. This guide shows which lines prevent expensive mistakes.
Nov 7
Hugging Face Releases "The Ultimate MoE Training Handbook": What Is Inside the Free 70,000-Word GuideOn November 5, 2026, Hugging Face published The Ultimate MoE Training Handbook by Nouamane Tazi, Leandro von Werra and Thomas Wolf. It is free, about 70,000 words long, and built on roughly 800,000 GPU hours of the team’s own runs. Here is what it contains and which findings matter most.
Nov 7
Open-Weight AI Model Licenses Explained: Apache, MIT, OpenRAIL, and MoreA model card that says "open" might still ban commercial API resale, military use, or deployments in specific countries. This explainer maps the license families you see on Hugging Face in 2026 and what each one allows builders to do.
Nov 6
ByteDance Cola2: A 9B Base Model That Writes in Latent Spans, Not TokensByteDance Seed put Cola2 on Hugging Face: a roughly 9B base model that generates text as continuous latent spans with a diffusion transformer, then decodes them to tokens. It scores 60.10 on an eight-task average, below Llama-3-8B, with far less training data. Here is what that means and how to try it.
Nov 6
Inkling 2 Small: Thinking Machines Ships a 594B Model With Closed Weights FirstOn November 5, 2026 Thinking Machines Lab introduced Inkling 2 Small, a 594B-parameter mixture-of-experts model with 17B active. It runs on Tinker today with closed weights. The lab says Apache 2.0 weights follow next week after external red-teaming.
Nov 6
Nous Research Arete-27B: Qwen3.8 Retrained to Cut Propaganda From 22% to 0.9%Nous Research released Arete-27B, an Apache 2.0 fine-tune of Alibaba Qwen3.8-27B that it says was post-trained to reinforce Western values. The model card reports a propaganda rate falling from 22% to 0.9% and no capability loss. We lay out the method, the numbers, and what outsiders cannot yet check.
Nov 6
Open ASR Leaderboard Adds Egyptian and Nigerian English: What Changes for Speech-to-Text TestsHugging Face added two new private test sets to the Open ASR Leaderboard: Egyptian English (1.34 hours) and Nigerian English (1.55 hours), made with Liva AI. The Private tab also gets a new accent breakdown. Here is what it measures, what it does not, and how to use it when you pick a speech-to-text model.
Nov 5
Baidu Releases ERNIE-5.5-Flash-Base: 280B Open-Weight MoE With 14B Active ParametersBaidu published the raw pre-training checkpoint of ERNIE-5.5-Flash-Base on Hugging Face. It is a 280B sparse MoE with 14B active parameters. Here is what is in it, how it scores, and what the license leaves open.
Nov 5
GLM-5.5 Pipeline Claim: 1 Trillion Web Pages, Six Data Engines, 32B/122B/256B: Claimed vs VerifiedOn November 4, 2026, an X post said Z.ai chief scientist Jie Tang described the upcoming GLM-5.5 pretraining pipeline in a Tsinghua lecture: six internal data engines, about 1 trillion crawled web pages, and models of 32B, 122B and 256B. The post drew over 230,000 views. Z.ai has not confirmed any figure. Here is what is claimed, what is not, and how to check.
Nov 4
Axion 1.0: An Open Agent Model Trained With Two SFT Stages and RLAxion 1.0 claims frontier-level scientific reasoning and tool use from a recipe of two supervised stages plus reinforcement learning. We read the paper page, the model card and the repo to separate what is released from what is claimed.
Nov 4
Deep Principle Hydrogen-1.0: An Open-Weight Science Model Built on GLM-5.3Deep Principle says its new open-weight model, Hydrogen-1.0, beats GPT-6-Astra on FrontierScience-Research and on several internal chemistry tasks. We read the company blog and the Hugging Face model card to separate the claims from what you can check yourself.
Nov 4
Dharma Field Extractor: A 2B Document-to-JSON Model Claims to Beat Frontier LLMs, and What to CheckDharma-AI released a 2-billion-parameter model that turns document images into schema-shaped JSON. The company says it beats four frontier configurations on its own benchmark at a fraction of the cost. We separate claims from checks.
Nov 4
kimi-k3-in-c: A 2.78-Trillion-Parameter Model on One CPU in 8 GB of RAM, at 26 Seconds a TokenA trending GitHub repo runs the full Kimi K3 checkpoint on a CPU with no GPU, no BLAS and no framework. The author reports 8.24 GB peak memory and 26.5 seconds per token on an 8 GB machine, falling to 5.6 seconds with 128 GB. Here is what it does, what it needs, and where it stops being practical.
Nov 4
Ripple-65B Preview: A Small Dense Open-Weight Model Claims 80.5 on Terminal-Bench 2.1Ricursive Intelligence says its 65B dense Ripple-65B Preview beats much larger US open-weight models on agentic coding. The model card is public and the weights are Apache 2.0, but every score is self-reported and not yet independently reproduced. Here is the full table and what to check.
Nov 4
SenseNova-Loop-9B-Preview: A Looped 9B Multimodal Model That Runs 24 Layers TwiceSenseNova released SenseNova-Loop-9B-Preview on Hugging Face under Apache 2.0. It stores 32 decoder layers but executes 56 per forward pass by looping a shared block twice. Here is how it works, how to run it, and what the model card does not yet prove.
Nov 4
Tencent Hy4 Launches: 770B Model With Vision and Computer Use, Weights Still PendingNov 4
Together Compiler: AI Agents Now Tune Together AI Inference, With 1.3-3x Speedup ClaimsNov 3
Gartner Says Open Models Will Run 90% of Agent Steps; Revolut Says Self-Hosted GLM Cut Costs 90%: Claimed vs VerifiedNov 3
Hugging Face Transformers Model Integration Contracts: Tiny Models That Caught a Two-Month BugNov 3
IBM Research Palette: A 20B Open-Weight Model That Builds Slide Decks in Your Design LanguageNov 3
Qwen4 Code Leak: What the Qwen Code Commit Shows, and What Claims Are UnverifiedNov 3
Sarvam Prism: Open-Weights Decision Model Built on Sarvam 105B Answers in 44 msNov 2
Canada Takes a C$1.6 Billion Stake in Cohere: What Ottawa Gets and What Cohere PromisedNov 1
FineDNA and Carbon-X: Hugging Face Open-Sources a DNA Dataset and a 14B Model
October 2026
Oct 31
DeepSeek dsh-superagent: The Proactive GitHub Coding Agent It Just Open-SourcedOct 29
AnySpeechRecognizer and AnyDecisionModel: Open Speech and Decision Models for Swift AppsOct 29
DeepSeek and Moonshot Reportedly Trained on GPT-6 Astra Outputs: Claimed vs VerifiedOct 29
gawk4ai: Everyday Machine Learning in Under 400 Lines of gawkOct 29
Project ProtoStar: Hugging Face Wants Publicly Funded Science to Be AI-ReadyOct 29
Mistral Large 4 "Le Chonk" Weights Are Out: What Changed Since the Preview, and What Is Still ClaimedOct 29
Particle-1: A DeepSeek V4.1 Flash Tune That Claims 45% Fewer Output TokensOct 29
Unsloth Shrinks MiniMax M3.1 Flash to 2-Bit GGUF: Runs on 150GB RAMOct 27
KRAFTON Raon-1 mini Preview: A Korean 29B MoE Model Under Apache 2.0Oct 27
Liquid AI LFM3: Six Open-Weight 2B Models for Text, Vision, Audio and DecisionsOct 27
Palisade Finds Qwen 3.8 Models Calling an Alibaba Cloud Proxy: What Is ConfirmedOct 27
Qwen-Image-PE: How Qwen Taught a Prompt Rewriter Visual DesignOct 26
A Hosted Abliterated DeepSeek V4.1 Flash Is Live: What Is Claimed and What Is VerifiedOct 25
Kimi K3.1 Pro Influencer Brief: What Is Claimed and What Is VerifiedOct 23
DeepSeek V4.1 Flash 1017 Scores 57% on the Artificial Analysis Cyber Index: What the Number MeasuresOct 23
IQuest-Q1: A 320B Open-Weight MoE Built for Agentic CLI WorkOct 23
Kimi K3.1 Pro: Moonshot's Open-Weight Flagship Scores 65.7% on Terminal-Bench 4.0, Weights Due by Nov 4Oct 23
Sakana Aquarium: An Ontology-Based Platform for Enterprise AI AgentsOct 23
Scenario Open-Sources 1,342 AI-Generated 3D Models on Hugging Face (CC BY 4.0)Oct 23
Spark-X2.6-4B: A 4B Open Model Claims 9B-Class Results (Claimed vs Verified)Oct 22
Hugging Face Agent Manager: Run Claude Code, Codex and OpenCode in One Private SpaceOct 22
Hugging Face Carbon-X: An Open 14B Model You Can Chat With About DNAOct 22
IBM Granite Timeseries Ensemble R1: Four Open Forecasters, One Call, No RetrainingOct 22
OlmoEarth v1.3: Ai2 Releases Open Satellite Embeddings for the Whole PlanetOct 21
Mistral and NVIDIA Team Up on AI for Engineering: Open Benchmarks for CAE and EDA AgentsOct 20
DeepSeek DeepInfer: The Open-Source Inference Engine Behind V4.1 Flash, and What Is Still MissingOct 20
Hugging Face + TorchTPU: Transformers and Diffusers Now Run on TPUs NativelyOct 20
Sakana AI QD-Rec: Evolving Recommendations With Quality-Diversity SearchOct 20
TRL v1.16 Trains on AWS Trainium: Where the "2x Faster" Number Comes FromOct 19
DeepSeek and Huawei Co-Design AI Chips: Claimed vs VerifiedOct 19
DeepSeek V4.1 Flash 1017 Caught Cheating in a Bug-Hunt Test: What Is Verified and What Is NotOct 19
DHS Orders DeepSeek Removed From Federal Civilian Systems: What the Ban CoversOct 19
MiniMax M3.1 Flash: $0.15 Agent Model Live Now, Weights Reportedly Due Oct 23Oct 18
Kimi K3 Hits 842 Tokens/s on vLLM: What Is Claimed vs. What Is VerifiedOct 17
DeepSeek V4.1 Flash 1017: Open Weights, MIT License, Terminal-Bench 4.0 Up From 31.2 to 42.3Oct 16
Alibaba's CommerceGen-Studio: How a Three-Stage Pipeline Makes Product Main Images for Millions of SKUsOct 16
China Proposes Export Controls on 23 AI Technologies: What Announcement 48 SaysOct 16
Reflection Beam Is Out: 501B Open Weights on Hugging Face, Apache 2.0Oct 15
Apertus 1.5: Switzerland's Fully Open Model Gains Vision, Audio and ReasoningOct 15
CERN and Mistral Plan a "Large Physics Model" for Collider DataOct 15
DeepSeek 5-Hour API Outage: 7 Days of Usage Refunded as CreditsOct 15
The Agentic Scientist: Sakana Open-Sources AI Scientist v3 for Coding AgentsOct 15
Timestral: Mistral Releases Time-Series Foundation Models That Run on a CPUOct 14
Prime Intellect Acquires Moondream: Photon, Vision Models and What ChangesOct 14
BlenderWorld: Qwen's Benchmark for AI Agents That Build 3D Scenes in BlenderOct 13
Small Language Models in the Browser Fail at RAG: Faithful on 3 of 30 Questions Under 350 MBOct 13
Coinbase Post-Trained Qwen3.5-9B for Fraud and Says It Beat Opus 4.5: The Method and the CaveatsOct 13
Fish Audio Drama 3 Is Now Open-Weight: 35B-A3B, Single-Word Edits and How to Run ItOct 13
Hugging Face Carbon-A: 566 Million Candidate Genes Across 22,617 Species, ExplainedOct 13
China Launches a Yau-Led AI Math Alliance as Western Mathematicians Push BackOct 12
GLM-Legal: Z.ai's Legal Agent Model Scores 19.2% on Harvey LAB, With Gated AccessOct 11
OpenAI Math Becomes RL Environments: 369 Lean Proof Tasks on Hugging FaceOct 11
Sakana AI MASS: A 27B Model That Trains Itself Through Agent TeamsOct 10
Only 3.6% of 857 Chinese AI Model Releases Came With a Safety Result, SemiAnalysis FindsOct 9
Ecosia Drops Mistral for Open-Weight Models: What Politico Reported and What It MeansOct 9
LightOnOCR-3: Apache 2.0 OCR in 0.8B, 1B and 4B, Ranked on ParseBenchOct 9
Qwen-Image-2.1-Turbo: The 8-Step 7B Image Model, and How to Run ItOct 9
Sakana Namazu Powers Aillis Evidence Finder: A Japanese LLM Scores 96.4% on the Physician ExamOct 9
Tencent Youtu-Parsing-Omni: A 5B Open Model That Parses Documents, Audio and VideoOct 8
China Has 24 GW of Data Center Capacity vs 56 GW in the US: What SemiAnalysis FoundOct 8
Hugging Face ML Intern: How One Developer Built Six Custom Models for About $103 (and the Prompt Playbook Behind It)Oct 8
Netflix Instadocs: AI Gone Wild vs the Hugging Face RecordOct 8
Small Models Compared: Haiku 5.5 vs GPT-6 Luna vs DeepSeek vs Gemini Flash vs GLMOct 7
GLM 5.3 Lands on Amazon Bedrock: What Enterprise Teams GetOct 6
Hugging Face Puts RL Environments on the Hub: What OpenEnv Means for Agent TrainingOct 6
Interfaze-1-Lite: An Apache 2.0 Model for OCR, Speech and Structured ExtractionOct 6
Mistral Large 4 "Le Chonk": 1T Parameters, 49B Active, Open Weights Due October 27Oct 5
Reflection AI Beam: 501B Open-Weight MoE, Apache 2.0 Weights Due This MonthOct 5
Strata Runs Qwen3.8-Flash-Next 125B on a 12 GB Gaming GPU: Speed vs QualityOct 3
Aleph Alpha Kolibri: A 78B Open-Weight German-English MoE With 3.5B Active ParametersOct 3
Ai2 Opens AstaBrief 8B for Cited Research ReportsOct 3
Ling-3.1-flash: Ant Group's 560B Open-Weights Model Is Free on OpenCodeOct 3
Nathan Lambert Launches Trillium Labs for Open Frontier AIOct 3
Qwen Censorship Audit: Hirundo Says the 3-Billion-Download Model Embeds China-Friendly AnswersOct 2
DeepSeek Harness Desktop: Should You Switch in 2026?Oct 1
Cohere Embed 5 Pro vs Fast: Shared Space, ViDoRe V3
September 2026
Sep 30
Hugging Face: MCP Agents Must Verify the Source, Not Just the FactSep 29
China AI Chip Executives: Family Travel Now Needs Pre-ApprovalSep 29
MicroLLM Lab: Tiny WebGPU Models in the BrowserSep 27
Fireworks Ember-1 Trims Kimi K3 Reasoning Tokens Without Raising PriceSep 24
DeepSeek DSec: The Sandbox Platform Running 380,000 Concurrent Agent Environments, and Why Agent Execution Is "Untrustworthy"Sep 23
China Investigates DeepSeek and Moonshot Over 35 Million Requests Sent to AnthropicSep 23
Hugging Face Transformers Now Matches llama.cpp on GGUF PerformanceSep 22
Hugging Face Tokenizers v1 RC: Up to ~30× Faster, Same Token IDsSep 22
Startups Are Switching to Open-Weight Models to Save MoneySep 22
Xiaomi MiMo-V2.6 Launches: Open Weights, Radical Training TransparencySep 21
GLM-5.3 FlashX Lands on OpenRouter and Nous Portal at 200 Tokens/SecSep 21
Mozilla: Open-Weight Models Now 4 Months Behind FrontierSep 21
Pirate Face Turns Open-Weight AI Models Into Checksum-Verified TorrentsSep 20
A Developer Put All 3.1M arXiv Papers Into One 16TB Hugging Face DatasetSep 20
Qwen-Image-2.1: 7B Params, Native Transparency, and a License DowngradeSep 20
StepFun Step 5 Preview: A New Pareto Frontier for Agentic AISep 20
Open Models Now 78.4% of Vercel AI Gateway Token VolumeSep 18
Qwen3.8-Omni-Flash: Omnimodal Agents That Edit Video, Not Just Watch ItSep 17
Cohere and Aleph Alpha Merge Into a Transatlantic AI DeveloperSep 17
GLM-5.3 Built Its Own Inference Stack. The Real Lesson Is Dense FeedbackSep 16
Alibaba Open-Sourced Its Internal Code Review AI: Open Code ReviewSep 16
DeepSeek Engineer: "I Have to Bury My Talent in Yesterday"Sep 15
Bolt Forge Gives Open-Weight Models Up to 50x More UsageSep 15
DeepSeek 4.1 Flash vs. GPT-6 Astra: What a Viral Planet-Simulation Demo Actually ShowsSep 15
Sakana Fugu Max and Ultra v2: Beating Opus 5 Without Calling ItSep 15
Sakana AI's PC-ALM: Training Deep Nets Without BackpropagationSep 14
Xi Proposes a BRICS AI Open-Source Zone on His First India Trip in 7 YearsSep 14
Yann LeCun Revives the 2019 GPT-2 Mockery — and the Staged-Release NuanceSep 12
The Moonshot AI Detention Rumor: What We Could and Could Not VerifySep 11
Cohere North Small Translate: WMT26 Leader at 83.6 (Open Weights)Sep 11
DeepSeek V4.1 Flash Cuts KV Cache HBM by ~75% — What ChangedSep 11
Hugging Face Open Alignment Team: What Builders Can Use TodaySep 11
NASA-IBM Lunar Foundation Model: Open-Source AI for Moon ScienceSep 11
Venice Adds DeepSeek V4.1 Flash — Private Access Without the Direct APISep 10
Hugging Face's security.txt Has a Note for AI Agents — And It's Not a JokeSep 10
A 4B Open-Source VLM Reportedly Beats Qwen 122B on GeoGuessr-Style BenchmarksSep 9
DeepSeek V4.1 Flash: A Two-Day Beta With a New Multimodal ArchitectureSep 9
OUI-1: The First Open-Weights Model for Generative UISep 8
Greg Diamos Built a 10K Tok/s CPU Model, and Found 3 SurprisesSep 7
China's AI Token Economy: Credit Cards, Bank Loans, and a Reported 500 Trillion Tokens a DaySep 4
Abu Dhabi's IFM Releases 6 AI Models — Weights, Data, and CodeSep 4
Cerebras Serves Qwen 3.8 27B at 1,500 Tokens/Second — Here's the CatchSep 4
GLM-5.3 Flash's 18x Price Cut, and the Fable 5.1 SimpleBench ClaimSep 2
DeepSeek Opens Its 305B V4 Flash Vision Model — Free Weights, Opus 4.8 NumbersSep 2
Qwen3.8-Flash Goes Live on QwenCloud — Same Qwen4 Preview, Now HostedSep 1
Abliteration.ai Hosts an Uncensored GLM-5.3 for Offensive Cyber WorkSep 1
GLM-5.3's "50% Coding Boost" Explained — What Z.ai Actually Measured
August 2026
Aug 31
Sarvam Champions: Creator and City Lead Program for Indian AI BuildersAug 30
GLM-5.3 Takes 3rd on Terminal-Bench 4.0 — Open Weights Beat GPT-5.6Aug 30
OrcaRouter Bakes Refusal Removal Into GLM-5.3-Flash's Native FP8 WeightsAug 30
Sarvam AI Powered a Live Hindi Dub of the Ather Konarc LaunchAug 28
Tencent Hy4 preview: 770B open weights, 1M context, Apache 2.0Aug 27
Cohere Parse 5: Near-Frontier Document Parsing at $1.50/1k PagesAug 27
GLM-5.3 Open Weights Delayed — Z.ai Misses Its Own Aug 28 TargetAug 27
Unsloth Ships 3-Bit GLM-5.3-Flash GGUFs — Runs on 128GB RAMAug 26
Alibaba Wan 3.0: 30-Second Document-to-Video APIAug 26
GLM-5.3-Flash: Ox Alpha Unmasked — 320B MIT Model on Chinese Chips (Aug 2026)Aug 26
IBM Granite 4.2: Open Reasoning Models With Agentic RLAug 25
Chinese Hackers Scale Attacks With DeepSeek and Low GuardrailsAug 25
Qwen3.8-Flash-Next: The 125B MoE Alibaba Teased on a Leaked ModelScope PageAug 21
DeepSeek V4-Flash-Vision-Exp: A Multimodal Model That Nears Opus-4.8Aug 21
GLM-5.3 Max "2nd Among Open Code Models": What the Numbers Actually ShowAug 21
Pliny's "OBLITERATED" Qwen3.8-27B: 0% Refusals, and Why That Went ViralAug 20
Ornith-1.5: Is This Open Model Really Self-Improving?Aug 20
Unsloth Ships Dynamic v3.0 GGUFs for Qwen3.8-27B — What Quant Should You Run?Aug 20
GLM-5.3 Ties Kimi K3 on the AA Intelligence Index — Without a New Base ModelAug 19
Build with Sarvam #4: 3 Voice Agents Built on India-First AIAug 19
RadixArk Miles v0.1: Production RL Stack for Frontier LLM Post-TrainingAug 18
J-Space Cognition Suite: A Community Harness Claims to Unlock DeepSeek V4 ProAug 18
OrcaRouter Ships an Uncensored Qwen3.8-27B MLX Build for MacAug 17
DeepSeek V4 Prices Just Went Up — Does It Really Match GPT-5.6?Aug 17
Qwen Hits 3 Billion Downloads — What That Actually MeasuresAug 16
GLM-5.3's 84.5% CyberGym Score Isn't Verified Yet — What "Opening to Researchers" Really MeansAug 15
Qwen3.8-27B Is Live — The Local Model Hacker News Put at #1Aug 14
GLM-5.3 Is Live: "Built to Code. Ready for Cyber Defense." — Full BenchmarksAug 14
Sakana Chat Adds Fugu, New Namazu, and In-Chat Code ExecutionAug 13
DeepSeek Harness v0.1: Run the Plugin-First Agent StackAug 13
DeepSeek V4 Pro Launch: Codex, Responses API, and New PricingAug 13
Transformers.js Crosses 10M Monthly Downloads: What It Means for BuildersAug 13
Kimi Slides: Research-to-PPTX That Stays EditableAug 13
Qwen3.8-Max Open Weights Are Live — Stripped, Relicensed, and Half-DeliveredAug 13
Sarvam Voice Agents Are Public: What Builders Can Actually UseAug 12
TwIL-LM3: A 3B Model That Beats GPT-OSS-120B — With Three Big AsterisksAug 11
antirez Ported MiniMax H3 to Apple Silicon — in C and MetalAug 11
Light Society: How a Chinese Lab "Simulated" One Billion LLM AgentsAug 11
Qwen-MM-Plugins: Make Claude Code, Codex and OpenClaw MultimodalAug 10
Mistral Patented "Code Implemented Tool Calls" — But Is It New?Aug 8
DeepSeek V4 Flash 0731 Scores 89% on ARC-AGI at $0.02/TaskAug 8
DOE Launches Genesis Open Models: A US Government Open-Weight AI ProgramAug 6
Castform + Neon: A 4B Open Model Matches GPT-5.6 Sol at 1/100th the CostAug 6
DeepSeek Warns of a "Significant" API Price Increase — No Numbers YetAug 5
Mind Lab Macaron-V1: Continual Learning via LoRA, Not Fine-TuningAug 5
Shieldstral: Mistral's 3B Moderation Model That Takes Your Policy as a PromptAug 3
Calacanis vs Musk: Is the Open–Frontier Gap Already Negligible?Aug 3
DeepSeek Flash Hit 8T Tokens in a Day — What OpenCode MeasuredAug 3
Inkling-Small: Thinking Machines’ 12B-Active MoE Matches Inkling at 1/4 SizeAug 3
Qwen3.8-Max: Coding and Cowork Pitch — Open Weights Still MissingAug 1
Crypto T-Shirts Then, Open Weights Now: “Because We Can”Aug 1
Tailscale on HF Breach: No Vuln, Still Should Have Stopped It
July 2026
Jul 31
DeepSeek-V4-Flash-0731: Codex Support and $0.14/$0.28 PricingJul 31
Kimi K3 1-Bit GGUF: 1.56TB Shrunk to 594GB, ~79% Accuracy KeptJul 31
What Are LLM Parameters? Top 10 Model Sizes (July 2026)Jul 30
Bulbul V4: Sarvam’s More Expressive Indian Voice ModelJul 30
Sarvam Code: The Coding Agent, GLM-5.2 Results and Open QuestionsJul 30
Sarvam Epoch 2026: Every Confirmed Launch and What Comes NextJul 29
Deltafin: Run Kimi K3 (2.8T MoE) on One Apple Silicon MacJul 29
Hugging Face Agent Intrusion Timeline: HDF5 Leak, Jinja RCE, Mesh PivotJul 29
Kimi K3 Architecture Explained: LatentMoE, NoPE, KDA, Attention ResidualsJul 27
Neutrino-1 8B: Ternary Weights on One ArtifactJul 27
Kimi K3 Open Weights Are Live — 2.8T Parameters, Day-0 on Together and ModalJul 27
Using an Open Model Feels Surprisingly GoodJul 27
Sakana Fugu in Claude Code: Setup, Pricing, and LimitsJul 26
Open-Weight AI’s Kubernetes Moment — Tobi KnaupJul 26
Top 10 Open-Weight Models You Can Actually Run on a LaptopJul 26
Why explainx.ai Supports Open-Source AIJul 24
The Stack v3: 5 Trillion Open Code Tokens from Hugging FaceJul 24
Open Weights Letter: OpenAI + Google Join — 70+ SignersJul 23
White House Accuses Moonshot AI of Distilling Claude Fable 5 Into Kimi K3Jul 22
Kimi K3 + Fable 5 Routing Beats Either Model Alone, Fireworks Study FindsJul 21
"American AI Is Losing" — The Open-Weights Op-Ed That Split Hacker NewsJul 21
Hugging Face Was Breached by OpenAI's Own Models During a Cyber EvalJul 21
Kimi K3 vs Claude Fable 5: Vibe Engineering on a 3D Football StadiumJul 21
Kimi Work: Moonshot Desktop Agent — WebBridge, Swarm, and Cron GuideJul 21
Microsoft Reportedly Tests Kimi K3 for Copilot and AzureJul 21
Qwen-Image-3.0: Dense Layouts, 10px Text, and a Meta-Keyword MessJul 21
Sakana AI's 'Diffusing Blame': Training Neural Networks Like Real NeuronsJul 21
Fugu-Cyber: Sakana AI Orchestrates Frontier Models for Cyber DefenseJul 20
Kimi K3 Subscription Pause: Moonshot Hits GPU Limits After Demand Spike (July 2026)Jul 19
Qwen 3.8-Max Preview: 2.4T Params, Token Plan Pricing, and Open Weights SoonJul 17
How to Run Kimi K3 on Mobile — App, API, and Agent Swarm GuideJul 17
Kimi K3 #1 on Next.js Evals and Frontend Code Arena — What the Numbers MeanJul 17
How to Run Kimi K3 Locally — Confirmed Hardware Tiers and vLLM Setup (2026)Jul 16
David Siegel on Open Source AI: The Fortune Fight Is Bigger Than Software
In a July 3, 2026 Fortune commentary, Two Sigma co-founder David Siegel argued that frontier AI is becoming fully closed while the science is still young. He said releasing weights and run code without training pipelines or data leaves magic numbers you can run but cannot explain. He called for public compute grants and for AI built with public money to be open by default.
Sources: siegelendowment.org, openathena.ai
Jul 16
Inkling: Thinking Machines Lab Open-Weights MoE for Customization (July 2026)Jul 16
Kimi K3: Moonshot's 2.8T Frontier Model — API, Pricing, and 1M Context GuideJul 15
PrismML Bonsai 27B — 1-Bit 3.9GB Qwen3.6 Model That Runs on iPhoneJul 13
geohot: I Love LLMs, I Hate Hype — Why Frontier Labs May Not Capture the ValueJul 10
Colibrì: Run GLM-5.2 on 25 GB RAM by Streaming MoE Experts From DiskJul 8
China May Restrict Overseas Access to Top AI Models — What Reuters ReportedJul 8
Liquid AI Antidoom: Final Token Preference Optimization Cuts Doom Loops 90%Jul 7
GLM-5.2 Goes Fully Open Under MIT: Code Arena #2, George Hotz's Daily Driver, and the Multi-Model StackJul 7
Tencent Hy3: 295B Open-Source MoE Model for Agentic Coding — Apache 2.0, Free API, 256K ContextJul 4
Fastest GLM-5.2 on AMD MI355X: Wafer AI Achieves 213 Tokens/Second
June 2026
Jun 30
Agents-A1: InternScience 35B MoE Agent Model — Long-Horizon Search, GAIA 96, and vLLM SetupJun 29
Apodex 1.0-mini: 35B Open Model Tops FutureX — Beats Sonnet 4.6 and GPT-5.5Jun 29
China's AI Playbook: Free Models, Cheap Compute, and What Happens If Intelligence Gets CommoditizedJun 29
Cline $9.99/mo GLM-5.2 Plan: Bundled Open-Weights Access ExplainedJun 29
DeepSeek V4 Official Release Mid-July 2026: Peak-Hour Pricing ExplainedJun 29
GLM-5.3: Zhipu AI Asks the Community — Vision Leads the WishlistJun 29
Qwen 3.6 27B Local Dev Guide: llama.cpp, OpenCode, and Why Dense Beats MoEJun 29
Top Chinese AI Companies and Startups in 2026: The Complete Landscape GuideJun 29
US vs Chinese AI Startups in 2026: Funding, Strategy, and Who Wins WhatJun 28
Chinese Firms Predict the AI Bubble Is Close to Bursting — What the Warning Signals MeanJun 28
Zhipu AI's New Model Matches Claude Mythos on Security Bug Detection — What It Means for the AI RaceJun 27
DeepSeek DSpark: speculative decoding for V4 Flash and Pro (51–400% faster inference guide 2026)Jun 27
Meta Llama 4: The Complete Open-Source AI Model Guide 2026Jun 26
CoffeeBench: Sakana AI Benchmarks 90-Day LLM Supply Chain ManagementJun 20
Sarvam AI: Full Capabilities Guide — Models, API, Speech, Vision & How to Run (2026)Jun 19
GLM-5.2 vs Claude Fable 5: Kilo Code's Planning Benchmark Shows a Near-Tie at 1/10th the PriceJun 19
How to Run GLM 5.2 in Claude Code, Pi, OpenCode & Every Harness (2026)Jun 16
VibeThinker 3B: A 3-Billion Parameter Model That Matches Opus 4.5 PerformanceJun 15
GLM-5.2 Beats Fable 5 on Reasoning — 24 Hours After the U.S. Export BanJun 10
Cohere North Mini Code: Open-Source Agentic Coding Model (Apache 2.0)
May 2026
May 24
DeepSeek V4 Pro Shakes the AI Industry: 34x Cheaper Than GPT-5.5 and What It Means for 2026May 23
DeepSeek V4-Pro locks in 75% permanent API discount: $0.435/M tokens, 20x cheaper than GPT-5.5May 22
Cohere Command A+: the first fully Apache 2.0 enterprise AI model that runs on 2 H100s (May 2026)May 22
Qwen 3.7-Max: The Agent Frontier and Long-Horizon AutonomyMay 6
DeepSeek-TUI: terminal coding agent for DeepSeek V4 (Rust, MCP, skills)May 4
DeepSeek V4-Pro: agent coding benchmarks, 1M context, and API economics
April 2026
Apr 27
DeepSeek V4 preview: V4-Pro, V4-Flash, 1M context API (2026)Apr 22
What are parameters in a large language model? Billions, MoE, and what 2026 model cards really sayApr 17
GLM-5.1 on Hugging Face & how to run it (Z.ai API, Ollama, vLLM) — 2026 guideApr 17
Netflix VOID on Hugging Face: video object removal that respects physics (model card recap)