Fine-Tuning

LoRA, RLHF, GRPO, model adaptation, training techniques

522 articles across 150 editions

Articles

  1. Claude Haiku 5.5 -- 2026-10-08
  2. reddit.com -- 2026-10-08
  3. Editorial video submission (YouTube: Cl3OWig5hkk) -- 2026-10-08
  4. AgentCyberRange: benchmarking frontier AI agents in realistic cyber ranges -- 2026-10-08
  5. AI Security Bootcamp (AISB): open curriculum repo for securing frontier AI systems -- 2026-10-08
  6. dealignai/GLM-5.3-CYBERSECURITY-FP8 (trending on Hugging Face) -- 2026-10-08
  7. [Editorial] RED-SNOW-5.3-FLASH EXL3 SAGE 2.49bpw quant (Hugging Face) -- 2026-10-07
  8. [Editorial] ViC305 on the RED-SNOW-5.3-FLASH EXL3 release (X post) -- 2026-10-07
  9. I quantized GLM-5.3-UNCENSORED to MXFP4 for AMD GPUs - weights available on Hugging Face -- 2026-10-07
  10. [Editorial] AliesTaha/fable-traces: compact Qwen3-4B instruct tune (Hugging Face) -- 2026-10-07
  11. Beam: Reflection's 501B open-weight model -- 2026-10-07
  12. [Editorial] The Bitter Lesson (Rich Sutton) -- 2026-10-07
  13. Qwen/Qwen-AgentWorld-35B-A3B -- 2026-10-07
  14. Several vulnerabilities have been discovered in the Linux kernel -- 2026-10-05
  15. [Editorial] Linux kernel commit 60d93c27 (torvalds/linux) -- 2026-10-05
  16. PatchBench: Evaluating AI Agents for Vulnerability Patching -- 2026-10-05
  17. [Editorial] Yukon: HashSmash -- 2026-10-05
  18. 500k facial scans at UK stations yield no arrests, 1 false positive -- 2026-10-01
  19. [Editorial] -- 2026-10-01
  20. [Editorial] -- 2026-10-01
  21. [Editorial] -- 2026-10-01
  22. reddit.com -- 2026-10-01
  23. [Editorial] Anthropic: GLM-5.3 and the spread of advanced cyber capabilities -- 2026-09-30
  24. [Editorial] FelonyBench: measuring whether AI agents escape their sandboxes and break the law -- 2026-09-30
  25. I Could've Accessed 17T Microsoft Records -- 2026-09-30
  26. PS5 Relapse Exploit -- 2026-09-30
  27. MLST: Can Rewriting an AI Agent Bend the Curve? (Weco self-modifying agent, 8 days) -- 2026-09-29
  28. The Biggest AI Coding Agent Upgrade Is Already on Your Machine?! (agent transcripts as data) -- 2026-09-29
  29. The AI Bottleneck: Why Your Team Isn't Shipping, and How to Fix It -- 2026-09-29
  30. [Editorial] Sebastian Raschka on LLM research (x.com/rasbt) -- 2026-09-24
  31. MiMo 2.6 Pro: Reducing overthinking and second-guessing -- 2026-09-24
  32. XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B -- 2026-09-24
  33. quants for K2-Horizon are now available -- 2026-09-24
  34. Claude Opus 5.5 -- 2026-09-23
  35. OpenAI GPT–6 Astra breaks Enigma message that has resisted solution since 2005 -- 2026-09-23
  36. Dream RSI (editorial pick) -- 2026-09-23
  37. WordPress core security advisory GHSA-7hp8-65ch-5whp -- 2026-09-23
  38. harisec: AI agent security research notes (gist) -- 2026-09-23
  39. Data-only attacks are easier than you think (2024) -- 2026-09-23
  40. PDF Forgeries Are Surprisingly Rare (2022) -- 2026-09-23
  41. LLM prompt injection testing at work just nuked our client demo and I feel sick -- 2026-09-23
  42. BLOOM-WILT: Logit Tilting for Behaviour Elicitation in Automated LLM Auditing -- 2026-09-22
  43. [Editorial] lordx64/phantom-kv -- 2026-09-22
  44. Frontier AI on Your Own Hardware -- 2026-09-22
  45. GLM Built Its Own Inference Infrastructure -- 2026-09-22
  46. I built a native Vulkan training backend for 143 modern Transformer architectures — no CUDA or PyTorch required -- 2026-09-22
  47. Jun Kim, oMLX creator and maintainer, joins Hugging Face to support the MLX community -- 2026-09-22
  48. [Editorial] Thom Wolf on X -- 2026-09-22
  49. Jonathan Taylor's detailed walkthrough (LinkedIn post by Kevin Kramer) -- 2026-09-21
  50. browser-use/jev-ultrafast: fast Jev runtime from the browser-use team -- 2026-09-21
  51. Post by @0xCodila on X -- 2026-09-21
  52. [Editorial] Hacktron: Hacking OpenAI -- 2026-09-18
  53. [Editorial] First agentic AI data breach reported to Spanish regulator -- 2026-09-18
  54. An Empirical Study of Harness Design for Coding Agents -- 2026-09-18
  55. [Editorial] Headroom: context optimization layer for AI agents (headroomlabs-ai/headroom) -- 2026-09-18
  56. Sub-agents burning your Claude Code 5-hour window? Check their 5-minute prompt cache -- 2026-09-18
  57. Discovered pi-vcc, why pi-blackhole? -- 2026-09-18
  58. [Editorial] forkd (deeplethe/forkd) on GitHub -- 2026-09-18
  59. Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher -- 2026-09-17
  60. [Editorial] Xiaomi MiMo 2.6 — Live Post-Training Dashboard -- 2026-09-17
  61. DeepSeek engineer relections on RSI - burying my talent to yesterday -- 2026-09-17
  62. UkisAI Swift-Qwen3.8-27B / -58.3% thinking, x1.95 speed while keeping the accuracy of xhigh -- 2026-09-16
  63. The new k2 horizon models seem like an absolute beast -- 2026-09-16
  64. Nex N2.5 Pro (407GB) released -- 2026-09-16
  65. openai/privacy-filter -- 2026-09-16
  66. seedleap/zing-world-model -- 2026-09-16
  67. Astra and Fable still hack on simple variants of alignment evals from 2025 -- 2026-09-14
  68. DS 4.1 and the new Harness -- 2026-09-14
  69. Should coding agents scan tool results before putting them into context? -- 2026-09-14
  70. decionis/docker -- 2026-09-14
  71. [Editorial] China-modified RTX 5090 with 96GB of memory appears on Alibaba for under $4,000 -- 2026-09-14
  72. Talk me out of buying a 3rd Spark -- 2026-09-14
  73. Dear 24G owners, try VLLM you might be able to run Qwen3.8 27B INT4, 144K FP8 KV on RTX 3090 with better speed. (TLDR VLLM AOT) -- 2026-09-14
  74. Nine coding harnesses vs. your laptop -- 2026-09-14
  75. After over a year of my nights and weekends, the Jenny app is done! -- 2026-09-14
  76. google/artemis -- 2026-09-11
  77. donvito/codex-astra-luna-orchestrator -- 2026-09-11
  78. [Editorial] -- 2026-09-11
  79. Kinesin movement powered by ATP (blender and unreal MCP) -- 2026-09-11
  80. Has anybody seen my keys? A key-hierarchy strategy for rack-level security -- 2026-09-11
  81. Does Runtime Topology Context Improve LLM-Generated Kubernetes Security Patches? -- 2026-09-11
  82. [Editorial] -- 2026-09-11
  83. [Editorial] Anthropic: Alignment assessment of cybersecurity incidents -- 2026-09-10
  84. [Editorial] Axios: Senate investigation into OpenAI and Hugging Face (Hawley) -- 2026-09-10
  85. [Editorial] YouTube video (eEBv0STiYhI) -- 2026-09-10
  86. Detecting and countering misuse of AI: September 2026 -- 2026-09-10
  87. Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra -- 2026-09-10
  88. Qwen3.8 Flash Next - Templates Comparison -- 2026-09-10
  89. Qwen 3.8 27b with PI agent - pushed to its 3D graphic game limits -- 2026-09-10
  90. [Editorial] YouTube: n9xKblqyQ28 -- 2026-09-09
  91. I ported Toyota's Lean quality system to Claude Code so the same agent mistakes stop coming back (MIT, free) -- 2026-09-09
  92. squall01337/mixamo-llm-mocap -- 2026-09-09
  93. Distributing Security Controls Through Harness Engineering -- 2026-09-09
  94. [Editorial] github/gh-aw: GitHub Agentic Workflows -- 2026-09-09
  95. Comprehensive Claude Code Permission Guard (settings.json) -- 2026-09-09
  96. Did anyone else notice that GPT-6/codex uses (inline) Python much more aggresively? -- 2026-09-09
  97. You Don't Need More VRAM To Run AI At Home (The Stack) -- 2026-09-08
  98. exllamav3 comfortably beats llama.cpp running CPU-offloaded Qwen-3.8-Flash-Next on my setup! -- 2026-09-08
  99. Spark-2.5-4B is an interesting model for 8GB Jetson Orin Nano Super SoC. -- 2026-09-08
  100. VoxGen, an AMD-optimized TTS inference engine for VoxCPM 2 models -- 2026-09-08
  101. Building a tiny ElevenLabs on a single 3090 in 2-hour runs. Here's the log of everything that broke. -- 2026-09-08
  102. [Editorial] GPT-6 Astra: OpenAI Deployment Safety Report -- 2026-09-04
  103. [Editorial] ARC Prize: Astra Evaluation -- 2026-09-04
  104. From 0-to-1 to 1-to-N: Reproducible Engineering Evidence for MetaAI Recursive Self-Design -- 2026-09-04
  105. Breaking Claude Code Opus 5 Auto Mode -- 2026-09-03
  106. C2PA manifest implementation warning -- 2026-09-03
  107. Show HN: FrontierHarness Eval – 9 harness, same model, cost per pass varies 17x -- 2026-09-02
  108. How to Build Agentic Graphs -- 2026-09-02
  109. llama.cpp GBNF grammars: constrained generation for local models -- 2026-09-02
  110. luvrix/fluxmend -- 2026-09-02
  111. Glitch-Cat-Club/graph-memory-starter -- 2026-09-02
  112. aka-rider/rune (GitHub) -- 2026-09-02
  113. ruvnet/ruClip (GitHub) -- 2026-09-02
  114. One note in three: a verified census of three deployed AI scribes, and the instrument that counted it -- 2026-09-01
  115. [Editorial] -- 2026-09-01
  116. Smartphone LED detects hidden cameras with AI -- 2026-09-01
  117. Open WebUI v0.11.1: The Model Learns to Stop and Ask -- 2026-09-01
  118. context-labs/whip -- 2026-09-01
  119. [Editorial] -- 2026-09-01
  120. Web Draw: drive a real browser from a text-only model, no vision required -- 2026-09-01
  121. Natural Ungrokking: Asymmetric Control of Which Rules Survive Pretraining -- 2026-08-31
  122. Claude Plays DOOM -- 2026-08-31
  123. Fully quantized NVFP4 Qwen3.8-27B with QUASAR QAD -- 2026-08-27
  124. Qwen3.8-Flash-Next better then DeepSeek V4 Pro -- 2026-08-27
  125. I really want DeepSeek V4 to work as a local coding agent, but the tool calling keeps falling apart. Has anyone solved this? -- 2026-08-27
  126. [Editorial] Whack-a-Mole Is Losing (David Adrian) -- 2026-08-26
  127. [Editorial] RFC 2119: Key Words for Indicating Requirement Levels -- 2026-08-26
  128. Claude subagent got bored and prompt injected my main session into deleting my database -- 2026-08-26
  129. actonos/actonos — a self-governing AI agent kernel running 24/7 -- 2026-08-26
  130. [Editorial] Agent Skill Creator -- 2026-08-26
  131. [Editorial] Editor's Working Notes (Google Doc) -- 2026-08-26
  132. I built a low-latency AI companion that plays Skyrim with me -- 2026-08-26
  133. Rimagination/h3lite — hardware-aware Codex skill for local MiniMax H3 video generation via ComfyUI -- 2026-08-26
  134. [Editorial] Scaling Memory Safety (Google Bug Hunters) -- 2026-08-25
  135. This Week in Security: Apple Warns Users, Stripe Merchants Leak Keys, Copilot Helps Hack Itself, and Comcast Senses Movement -- 2026-08-25
  136. [Editorial] gods-eye-view -- 2026-08-25
  137. [Editorial] Video Feature -- 2026-08-25
  138. Simon Willison: Qwen 3.8 27B is excellent, but it defaults to wildly overthinking things -- 2026-08-19
  139. Ollama's MTP variant of Qwen3.8-27B is 2x slower than the non-MTP one, measured, with a negative control -- 2026-08-19
  140. Taking Qwen3.5-9B quants to SOTA. New lineup incoming :) -- 2026-08-19
  141. [Paper] Intern-S2-Mobius: Foundation Model with Decoupled Knowledge and Reasoning -- 2026-08-19
  142. dealignai/Gemma-4-31B-JANG_4M-CRACK -- 2026-08-19
  143. Meta releases open weights for Muse Glimmer-30B -- 2026-08-13
  144. Qwen3.6 35B (2 min) vs Muse Glimmer 30B (4 min) on custom Llama.cpp build (RTX 5080) -- 2026-08-13
  145. Qwen 35B-A3B MoE vs 27B dense in local coding tests: ~4× faster, much smaller quality gap than I expected -- 2026-08-13
  146. I asked DeepSeek-V4-Flash to work with Muse-Glimmer for Vision ability in PI agent and it produced this -- 2026-08-13
  147. enabling PCI-E p2p for consumer Nvidia cards will yield you more than you think -- 2026-08-12
  148. Rumored 50-series Super refresh bumps everything +50% VRAM -- 2026-08-12
  149. Muse Glimmer 30B + DFlash speculative decoding on vLLM: 6 patches needed, 25 → 57 tok/s. Dockerfile and numbers inside. -- 2026-08-11
  150. Apple Silicon and macOS VMs: 11–16× Faster LLM Inference with Llama.cpp -- 2026-08-11
  151. Echo Dot 2 can run 28M LLM at decent speed -- 2026-08-11
  152. Chunked KL loss for running Knowledge Distillation locally (<6GB VRAM at 32K context length) -- 2026-08-11
  153. Sonic Pi v5 -- 2026-08-11
  154. Harness Engineering for Self-Improvement -- 2026-08-04
  155. Dual-Dimensional Consistency: Balancing Budget and Quality in Adaptive Inference-Time Scaling -- 2026-08-04
  156. [Paper] Towards Scalable Lifelong Knowledge Editing with Selective Knowledge Suppression -- 2026-08-04
  157. Pydantic AI 2.0: The New Best Way to Build AI Agents is Composing Capabilities -- 2026-07-29
  158. [Editorial] The Gauntlet Loop -- 2026-07-29
  159. The Blind Curator: How a Biased Judge Silently Disables Skill Retirement in Self-Evolving Agents -- 2026-07-29
  160. [Editorial] Anatomy of a Frontier Lab Agent Intrusion: Full Technical Timeline -- 2026-07-29
  161. [Editorial] HuggingFace Agent Intrusion: Interactive Replay -- 2026-07-29
  162. Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident -- 2026-07-29
  163. Document-borne AI worms can self-propagate through Copilot for Word -- 2026-07-29
  164. Kimi-K3 Releases on HuggingFace 6/27 -- 2026-07-28
  165. Ling-3.0-flash weights: SGLang says day-0, vLLM says when they land, llama.cpp closed the 2.6 request as not_planned -- 2026-07-28
  166. Sanctions on Open Source. hope they don't do anything stupid here. -- 2026-07-28
  167. [Editorial] -- 2026-07-28
  168. [Editorial] -- 2026-07-28
  169. I distilled an 8B teacher into a 0.6B student on my Mac (MLX). The 0.6B went from 36% to 100% on the task, but few-shot prompting actively made it worse. -- 2026-07-28
  170. OpenAI and Anthropic unite against open-weight AI risks to their bottom line -- 2026-07-27
  171. [Editorial] -- 2026-07-27
  172. [Editorial] -- 2026-07-27
  173. What do we know about the "AI Accelerators" used to train LongCat-2? -- 2026-07-24
  174. Running a 13M ASR conformer on a microcontroller -- 2026-07-24
  175. Running Qwen 3.6 35B MoE (Q4_K_M) on a Zeus (Xiaomi 12 Pro, 12GB RAM) -- 2026-07-24
  176. Squeeze-Release: Iterative Pruning with Exact Structural Minimization -- 2026-07-24
  177. [Editorial] -- 2026-07-23
  178. CEO of Hugging face: Heading to San Francisco to have a little chat with that "rogue agent" -- 2026-07-23
  179. [Editorial] -- 2026-07-21
  180. [Editorial] -- 2026-07-21
  181. Kimi K3 Agentic Benchmark -- 2026-07-17
  182. Kimi k3 is 2.8t! Will need to have an aggressive iQ2_XXS or IQ1.8! -- 2026-07-17
  183. Governments, companies, nonprofits should invest in free, open source AI [pdf] -- 2026-07-17
  184. Zig Creator Calls Spade a Spade, Anthropic Blows Smoke -- 2026-07-15
  185. Apple sues OpenAI alleging trade secret theft, says scheme was 'at every level' -- 2026-07-15
  186. China's DeepSeek developing its own AI chip, sources say -- 2026-07-15
  187. Data centers have hiked electricity prices on the public by $23B -- 2026-07-15
  188. [Editorial] CMMC Baseline Did Not Move — The Floor Did -- 2026-07-14
  189. Anthropic warns that AI will soon be able to improve itself without human intervention -- 2026-07-13
  190. Zhipu founder backs open-source AI as global security debate intensifies -- 2026-07-13
  191. [Editorial] -- 2026-07-13
  192. Kode Dot Programmable pocket device for makers, pentesters and geeks -- 2026-07-13
  193. stabilityai/stable-audio-3-medium -- 2026-07-13
  194. LiquidAI/LFM2.5-230M -- 2026-07-13
  195. Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents -- 2026-07-10
  196. [Editorial] -- 2026-07-10
  197. Getting Someone Else's Chat -- 2026-07-10
  198. [Editorial] -- 2026-07-10
  199. rheodev/cpa-plugin-privacyfilter -- 2026-07-10
  200. The Underhanded C Contest -- 2026-07-03
  201. Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers -- 2026-07-02
  202. New bench designed for smaller models: ObviousBench.com -- 2026-07-02
  203. I built an autonomous dev pipeline and ran the same project head to head: a 27B local on a modded 4090, then again on cheap cloud LLMs -- 2026-07-02
  204. Stdlib or Third-Party? Empirical Performance and Correctness of LLM-Assisted Zero-Dependency Python Libraries -- 2026-07-02
  205. DeepSeek V4 official version launching mid-July -- 2026-06-30
  206. OpenPangu-2.0-Flash: 92B MoE (6B active) on Ascend with 512K context -- 2026-06-30
  207. Anthropic's Amodei: "Open Source models [could take us to] a very dangerous place." -- 2026-06-30
  208. Even Google still believes in small models for coding — Gemma 4 31B hackathon at 1500 tok/s -- 2026-06-30
  209. Previewing GPT‑5.6 Sol: a next-generation model -- 2026-06-29
  210. [Editorial] -- 2026-06-29
  211. [Editorial] -- 2026-06-29
  212. We built a calibration-aware Q4_K_M quant of Qwen3.5 0.8B that recovers 96.5% of the BF16 gap vs pure llama.cpp Q4_K_M (SpectralQuant) -- 2026-06-29
  213. Update: First Manual Results from Testing Procedural Skill Transfer in Small Models -- 2026-06-29
  214. Multi Tier MoE Caching -- 2026-06-29
  215. GLM-5.2 is a step change for open agents -- 2026-06-25
  216. poolside/Laguna-M.1 · Hugging Face - 225B-A23B -- 2026-06-25
  217. Mimo 2.5 is _fast_ at large context (dual RTX Pro 6000) -- 2026-06-25
  218. The Eagle(3) has landed (for Qwen) -- 2026-06-25
  219. CPU-only TTS benchmark: Kokoro 82M vs Supertonic 3 vs Inflect-Nano-v1 (4.6M params), with UTMOS scoring on every sample -- 2026-06-25
  220. [Editorial] Video Content -- 2026-06-16
  221. [Editorial] Video Content -- 2026-06-16
  222. [Editorial] Video Content -- 2026-06-16
  223. [Editorial] StandardAgents Arrow-JS — JavaScript Agent Framework -- 2026-06-16
  224. archex: Local-First Deterministic Code-Context for AI Agents — No API Key, No Telemetry (Apache 2.0) -- 2026-06-16
  225. Ironsmith: Open Source macOS App That Creates macOS Apps From Prompts — Works With Local Models -- 2026-06-16
  226. zengxiao-he/tessera -- 2026-06-11
  227. Tencent-Hunyuan/UniRL -- 2026-06-11
  228. SafeAdapt: Provably Safe Policy Updates in Deep Reinforcement Learning -- 2026-06-11
  229. Physics-Grounded Multi-Agent Architecture for Traceable, Risk-Aware Human-AI Decision Support in Manufacturing -- 2026-06-11
  230. Config Files That Run Code: Supply Chain Security Blindspot -- 2026-06-10
  231. Surveillance is not safety: A statement on the UK's latest threat to privacy -- 2026-06-10
  232. FrontierCode -- 2026-06-10
  233. Introducing North Mini Code: Cohere's First Model For Developers -- 2026-06-10
  234. [Editorial] jedArden/ARMOR -- 2026-06-05
  235. m-sec-org/wafkiller -- 2026-06-05
  236. zmn-hamid/sni-spoofing-scanner -- 2026-06-05
  237. Everything at Every Scale: Scale-Invariant Diffusion with Continuous Super-Resolution -- 2026-06-04
  238. Gaussian Point Splatting -- 2026-06-04
  239. DaVinci Resolve 21 -- 2026-06-04
  240. Profiling in PyTorch (Part 1): A Beginner's Guide to torch.profiler -- 2026-06-04
  241. microsoft/harrier-oss-v1-270m -- 2026-06-04
  242. CohereLabs/tiny-aya-global -- 2026-06-04
  243. Use your Nvidia GPU's VRAM as swap space on Linux -- 2026-06-03
  244. [Editorial] chipotlai-max -- 2026-06-03
  245. [Editorial] Video Submission -- 2026-06-03
  246. It is an amazing time for programmers -- 2026-06-03
  247. A 10 year old Xeon is all you need -- 2026-06-02
  248. Inferencing at 10.33 t/s on Qwen 3.5 35B on a $300 laptop -- 2026-06-02
  249. OSC: Hardware Efficient W4A4 Quantization via Outlier Separation in Channel Dimension -- 2026-06-02
  250. Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillation -- 2026-06-02
  251. Qwen3.6 huge quality gain from Q4 to Q6 for coding agent -- 2026-06-02
  252. 1-Bit Bonsai Image 4B Image Generation for Local Devices -- 2026-06-01
  253. Old Mac Pro still proving its worth -- 2026-06-01
  254. Heterogeneous GPU Weighting & Layer Splitting -- 2026-06-01
  255. I finally put my NPU (Intel Arrow Lake) to use doing ASR for my smart home -- 2026-06-01
  256. OpenMOSS-Team/MOSS-TTS-v1.5 · Hugging Face -- 2026-06-01
  257. [Editorial] -- 2026-06-01
  258. [Editorial] -- 2026-06-01
  259. [Editorial] -- 2026-06-01
  260. [Editorial] -- 2026-06-01
  261. [Editorial] Barracuda Nightmare Eclipse Zero-Days -- 2026-05-29
  262. [Editorial] Video -- 2026-05-29
  263. Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL -- 2026-05-28
  264. PACZero: PAC-Private Fine-Tuning of Language Models via Sign Quantization -- 2026-05-28
  265. If AI writes your code, why use Python? -- 2026-05-12
  266. [Editorial] Lex Fridman — DeepSeek Deep Dive with Dylan Patel & Nathan Lambert -- 2026-05-12
  267. [Editorial] Reuven Cohen on AI Industry Developments -- 2026-05-12
  268. [Editorial] When Helpfulness Becomes Sycophancy -- 2026-05-08
  269. [Editorial] Vibe-Cast JEPA — Joint Embedding Predictive Architecture Exploration -- 2026-05-08
  270. [Editorial] -- 2026-05-07
  271. ProgramBench: Can we really rebuild huge binaries from scratch? (doesn't look like it) -- 2026-05-07
  272. Adding Benchmaxxer Repellant to the Open ASR Leaderboard -- 2026-05-07
  273. Does the "6 months gap" still hold? -- 2026-05-07
  274. Claude Code @ Opus 4.7 vs OpenCode @ qwen3.6:27b. Both shipped a playable cozy roguelite. -- 2026-05-07
  275. Fine-tuned Qwen3.6-35B-A3B DeltaNet experiment -- 2026-05-07
  276. Zyphra/ZAYA1-8B -- 2026-05-07
  277. Anthropic ships Claude for Creative Work with nine MCP-native connectors -- 2026-05-05
  278. n8n Just Got a New Tool (and it can SUPERCHARGE Claude Automations) -- 2026-05-05
  279. [Editorial] How to Use ADRs in Ruflo -- 2026-05-05
  280. Microsoft and OpenAI end their exclusive and revenue-sharing deal -- 2026-05-01
  281. [Editorial] Video: AI Development Insights -- 2026-05-01
  282. Talkie: a 13B vintage language model from 1930 -- 2026-05-01
  283. Your phone is about to stop being yours -- 2026-04-29
  284. OpenAI almost banned me because I tried to automate YouTube download -- 2026-04-29
  285. [Editorial] Video editorial submission -- 2026-04-29
  286. GPT-5.5 is out -- 2026-04-28
  287. OpenAI CEO's Identity Verification Company Announced Fake Bruno Mars Partnership -- 2026-04-28
  288. Trump picked a fight with Anthropic. Now the administration is backing off. -- 2026-04-28
  289. [Editorial] -- 2026-04-28
  290. Generalization at the Edge of Stability -- 2026-04-28
  291. [Editorial] -- 2026-04-28
  292. [Editorial] NIST Cybersecurity MLX Pipeline -- 2026-04-27
  293. [Editorial] LLM Fine-Tuning Guide -- 2026-04-27
  294. [Editorial] Open Generative AI — Curated Resource List -- 2026-04-27
  295. [Editorial] SecWest: AMD VVI — Hardware-Level Vulnerability Research -- 2026-04-20
  296. [Editorial] Phenoelit/Halvar — Legendary Security Research -- 2026-04-20
  297. Lean proved this program correct; then I found a bug -- 2026-04-14
  298. Multi-Agentic Software Development Is a Distributed Systems Problem -- 2026-04-14
  299. [Editorial] The Room That Quoted Back -- 2026-04-14
  300. [Editorial] -- 2026-04-13
  301. Qwen3.5-397B is shockingly useful at Q2 -- 2026-04-13
  302. [Editorial] -- 2026-04-13
  303. Liquid AI releases LFM2.5-VL-450M - structured visual understanding at 240ms -- 2026-04-13
  304. LLM Novice Uplift on Dual-Use Biology Tasks — 4x Accuracy Boost Bypasses Safeguards -- 2026-04-10
  305. [Editorial] Your AI Is Developing Capabilities Nobody Tested -- 2026-04-10
  306. The current state of the Chinese LLMs scene -- 2026-03-26
  307. Alibaba confirms they are committed to continuously open-sourcing new Qwen and Wan models -- 2026-03-26
  308. Cursor's Composer 2 apparently built on Kimi K2.5 without attribution -- 2026-03-26
  309. Nemotron Cascade 2 30B A3B -- 2026-03-26
  310. NVIDIA 2026 Conference LIVE. New Base model coming! -- 2026-03-20
  311. Nemotron 3 Nano 4B: A Compact Hybrid Model for Efficient Local AI -- 2026-03-20
  312. Granite 4.0 1B Speech: Compact, Multilingual, and Built for the Edge -- 2026-03-20
  313. [New Model & Agent] LocoTrainer-4B: A Claude Code-style local agent designed specifically to master the MS-SWIFT framework (4B, 32K, GGUF) -- 2026-03-20
  314. shallowdream204/BitDance-14B-16x -- 2026-03-20
  315. [Editorial] Karpathy autoresearch -- 2026-03-10
  316. [Editorial] Doc-to-LoRA -- 2026-03-10
  317. GPT-5.4 -- 2026-03-09
  318. [Editorial] -- 2026-03-09
  319. YuanLabAI/Yuan3.0-Ultra: 1010B MoE, fully open weights -- 2026-03-05
  320. We could be hours (or less than a week) away from true NVFP4 support in Llama.cpp GGUF format -- 2026-03-05
  321. Step-3.5-Flash-Base & Midtrain (in case you missed them) -- 2026-03-05
  322. Qwen3.5-9B Uncensored Aggressive Release (GGUF) -- 2026-03-05
  323. unknown -- 2026-03-05
  324. [Editorial] David Maynor Security Gist -- 2026-03-04
  325. [Editorial] arXiv:2602.23093 -- 2026-03-04
  326. unpromptedcon.org -- 2026-03-04
  327. Inside the M4 Apple Neural Engine, Part 1: Reverse Engineering -- 2026-03-03
  328. Hydroph0bia – fixed SecureBoot bypass for UEFI firmware from Insyde H2O (2025) -- 2026-03-03
  329. [Editorial] -- 2026-02-26
  330. [Editorial] -- 2026-02-26
  331. [Editorial] -- 2026-02-26
  332. Free ASIC Llama 3.1 8B inference at 16,000 tok/s - no, not a joke -- 2026-02-25
  333. [Editorial] Cognitum -- 2026-02-25
  334. Hetzner Prices increase 30-40% -- 2026-02-25
  335. [Editorial] -- 2026-02-24
  336. Anthropic Accuses DeepSeek, Moonshot AI, and MiniMax of Creating 24,000 Fake Claude Accounts -- 2026-02-24
  337. Gemini 3.1 Pro -- 2026-02-23
  338. 15 years of FP64 segmentation, and why the Blackwell Ultra breaks the pattern -- 2026-02-20
  339. Nvidia and OpenAI abandon unfinished $100B deal in favour of $30B investment -- 2026-02-20
  340. Google releases Gemini 3.1 Pro with Benchmarks -- 2026-02-20
  341. praetorian-inc/brutus -- 2026-02-19
  342. [Editorial] Unicornscan Getting Started -- 2026-02-19
  343. [Editorial] Unicornscan Alicorn -- 2026-02-19
  344. What Your Bluetooth Devices Reveal About You -- 2026-02-19
  345. [Editorial] When everyone can build software, who learns well? -- 2026-02-19
  346. Sonnet 4.6 feels like Opus 4.5 at Sonnet pricing -- 2026-02-19
  347. Anthropic Raises $30,000,000,000 As Run-Rate Revenue Grew 10x Annually Over Three Years -- 2026-02-19
  348. REASONING AUGMENTED RETRIEVAL (RAR) is the production-grade successor to single-pass RAG -- 2026-02-19
  349. Qwen Released Qwen 3.5 397B and Qwen 3.5 Plus! -- 2026-02-17
  350. Qwen3.5 NVFP4 (Blackwell) is up! -- 2026-02-17
  351. Running Gemma 3n E2B natively on Android via LiteRT -- 2026-02-17
  352. Deploying Open WebUI + vLLM on Amazon EKS -- 2026-02-17
  353. [Editorial] https://forge-quality.dev/articles/case-of-passing-tests-investigation -- 2026-02-02
  354. [Editorial] https://www.linkedin.com/posts/ownyourai_deepseek-just-released-the-first-vision-ai-activity-7421818927657385987-V1yo -- 2026-01-27
  355. Unsloth announces support for finetuning embedding models -- 2026-01-27
  356. matrixhub-ai/matrixhub -- 2026-01-14
  357. HM-RunningHub/ComfyUI_RH_DreamID-V -- 2026-01-14
  358. YouTube has removed the ability to search by upload date -- 2026-01-14
  359. Tried this open-source framework for LLM fine-tuning over UI -- 2025-12-12
  360. Golang optimizations for high‑volume services -- 2025-12-12
  361. [Editorial] https://www.linkedin.com/posts/stuart-winter-tear_assessing-llms-for-serendipity-discovery-activity-7396596796938153984-JY9u -- 2025-12-12
  362. Deprecations via warnings don't work for Python libraries -- 2025-12-11
  363. The "Confident Idiot" Problem: Why LLM-as-a-Judge fails in production. -- 2025-12-10
  364. Toyota unintended acceleration and the big bowl of "spaghetti" code (2013) -- 2025-12-09
  365. Free yourself from the Spotify desktop client with spotifyd -- 2025-12-04
  366. Llamacpp Parameters Tuning -- 2025-12-02
  367. [Editorial] https://huggingface.co/blog/grimjim/norm-preserving-biprojected-abliteration -- 2025-12-02
  368. Pong Gets the Boot -- 2025-11-26
  369. Building the largest known Kubernetes cluster, with 130k nodes -- 2025-11-26
  370. The Qtile Window Manager: A Python-Powered Tiling Experience -- 2025-11-25
  371. Most Stable Raspberry Pi? Better NTP with Thermal Management -- 2025-11-25
  372. Quantum physicists have shrunk and "de-censored" DeepSeek R1 -- 2025-11-20
  373. Gain 60% performance on RDNA 4 using this fix -- 2025-11-19
  374. Scale-out is the silent killer of LLM applications. Are we solving the wrong problem? -- 2025-11-19
  375. [Editorial] https://brianhorakh.medium.com/just-mcp-to-reduce-context-waste-in-spec-driven-development-3935922da5cf -- 2025-11-18
  376. [AutoBE] Qwen3-80B suddenly wrote doomsday AI mythology while generating a TODO app -- 2025-11-18
  377. My trick for better Claude Code collaboration: CLAUDE.md with conditional loading -- 2025-11-18
  378. A proper way to connect a local LLM to iMessage? -- 2025-11-13
  379. How do I level up from normie to normie pro with Claude -- 2025-11-13
  380. POC: Model Context Protocol integration for native Ollama app -- 2025-11-12
  381. Skills are in a weird middle ground between RAG and Custom GPTs, and I think that's why they feel so awkward -- 2025-11-12
  382. Native LLM Router Integration with Cost Transparency for OpenWebUI -- 2025-11-12
  383. Last week in Multimodal AI - Local Edition -- 2025-11-12
  384. DeepSeek-OCR GGUF model runs great locally - simple and fast -- 2025-11-12
  385. Qwen3-VL works really good with Zoom-in Tool -- 2025-11-12
  386. lightonai/LightOnOCR-1B-1025 -- 2025-11-12
  387. Qwen/Qwen3-VL-2B-Thinking -- 2025-11-12
  388. [Editorial] https://www.linkedin.com/posts/daniel-cuthbert0x_a-month-ago-gadi-evron-and-i-set-about-building-ugcPost-7393643597729845248-TSTD -- 2025-11-11
  389. Breakdown of New RunC Vulnerabilities -- 2025-11-11
  390. [Editorial] https://www.linkedin.com/posts/andriyburkov_this-paper-shows-a-27-million-parameter-model-activity-7393432619365052416-SFLO -- 2025-11-10
  391. Trajectory Distillation for Foundation Models -- 2025-11-10
  392. sail-sg/Precision-RL -- 2025-11-10
  393. inclusionAI/LLaDA2.0-flash-preview -- 2025-11-10
  394. Finetuning DeepSeek 671B locally with only 80GB VRAM and Server CPU -- 2025-11-07
  395. IPEX-LLM llama.cpp portable GPU and NPU working really well on laptop -- 2025-11-07
  396. Building a PV Solar-Powered Quadcopter -- 2025-11-07
  397. Adding a RTX 5080 into a 2U server with OcuLink -- 2025-11-06
  398. Why does Image Recognition work in llama-server but not through Open WebUI? -- 2025-11-06
  399. 2025 Component Abuse Challenge: A Piezo Disk Powers A Transmitter -- 2025-11-06
  400. [Editorial] Does the EU know that there are many countries outside of the EU that do not care at all about their -- 2025-11-03
  401. Ilya Sustkever's deposition reveals previously unknown details [pdf] -- 2025-11-03
  402. CISA and NSA share tips on securing Microsoft Exchange servers -- 2025-11-02
  403. The Smol Training Playbook: The Secrets to Building World-Class LLMs -- 2025-11-02
  404. Latest Update from Anthropic's new model - Neptune V6 -- 2025-11-02
  405. AI "Phone Farm" Startup Gets Funding from Marc Andreessen to Flood Social Media With Spam -- 2025-11-02
  406. Minimax-M2 cracks top 10 overall LLMs (production LLM performance gap shrinking: 7 points from GPT-5 in Artificial Analysis benchmark) -- 2025-11-01
  407. 🚨 OpenAI Gives Microsoft 27% Stake, Completes For-Profit Shift -- 2025-11-01
  408. FlashPack: High-throughput tensor loading for PyTorch -- 2025-11-01
  409. Kafka is Fast – I'll use Postgres -- 2025-11-01
  410. Analog Surround Sound Was Everywhere, But You Probably Didn’t Notice -- 2025-11-01
  411. Optimizing gpt-oss-120B on AMD RX 6900 XT 16GB: Achieving 19 tokens/sec -- 2025-10-31
  412. Flamingo 3 released in safetensors -- 2025-10-31
  413. Jeep Issues Emergency Recall for OTA-Bricked Wrangler 4xes -- 2025-10-29
  414. queenkiley/AI-Art-Generator -- 2025-10-26
  415. Unlock the power of images with AI Sheets -- 2025-10-26
  416. Open WebUI Context Menu -- 2025-10-26
  417. [Editorial] Browsers you can socially engineer -- 2025-10-24
  418. Update on Plans for Privacy Sandbox Technologies -- 2025-10-24
  419. PlayDiffusion finetune for audio inpainting non-verbal tags -- 2025-10-21
  420. Nvidia has produced the first Blackwell wafer on US soil -- 2025-10-21
  421. Show HN: Syna – Minimal ML and RL Framework Built from Scratch with NumPy -- 2025-10-21
  422. C Project Turns Into Full-Fledged OS -- 2025-10-21
  423. [Editorial] Agentic Orchestration -- 2025-10-19
  424. Chicken Squisher 3000: Squish-Proof Security -- 2025-10-19
  425. [Editorial] Claude Skills are awesome, maybe a bigger deal than MCP -- 2025-10-18
  426. [Editorial] Claude Skills -- 2025-10-18
  427. Copy-and-Patch: A Copy-and-Patch Tutorial -- 2025-10-17
  428. We built 3B and 8B models that rival GPT-5 at HTML extraction while costing 40-80x less - fully open source -- 2025-10-17
  429. Comparing Popular AI Evaluation Platforms for 2025 -- 2025-10-17
  430. State of AI Report 2025 -- 2025-10-17
  431. Nvidia breakthrough gives 4-bit pretraining technique the accuracy of FP8 -- 2025-10-15
  432. AI assisted suite - Doubt about n_gpu layer test -- 2025-10-15
  433. ibm-granite/granite-4.0-h-micro -- 2025-10-15
  434. Qwen/Qwen3-VL-235B-A22B-Instruct -- 2025-10-15
  435. Get your VLM running in 3 simple steps on Intel CPUs -- 2025-10-15
  436. A 5-minute, no-BS way to pick a local model for your real task -- 2025-10-14
  437. [Update] CodeLens.AI - Crowdsourced AI Leaderboard 3 Days Later: Blind Voting and What We Learned -- 2025-10-14
  438. ZephrFish/OmniProx -- 2025-10-14
  439. Preference optimization with ORPO and LoRA -- 2025-10-12
  440. [Show] SpiralTorch: A Rust-based PyTorch-style autograd engine (Python 3.14-ready) -- 2025-10-12
  441. 2G Gone? Bring It Back Yourself! -- 2025-10-12
  442. [Editorial] https://www.anthropic.com/research/small-samples-poison -- 2025-10-11
  443. [Editorial] https://www.linkedin.com/pulse/from-chatbot-operating-system-what-openais-next-move-means-leimer-ju18c -- 2025-10-11
  444. Rubygems.org AWS Root Access Event – September 2025 -- 2025-10-11
  445. Stop flexing Pass@N — show Pass-all-N -- 2025-10-11
  446. Architecting a project for optimal AI coding, any tips? -- 2025-10-11
  447. Basekick-Labs/arc -- 2025-10-11
  448. ServiceNow-AI/Apriel-1.5-15b-Thinker -- 2025-10-11
  449. meituan-longcat/LongCat-Flash-Chat -- 2025-10-11
  450. Did anyone try out GLM-4.5-Air-GLM-4.6-Distill ? -- 2025-10-10
  451. Thank you Anthropic &amp; this community! Our little side project just hit 1M visits and even made it on National TV! -- 2025-10-10
  452. Sharing my free tool for easy handwritten fine-tuning datasets! -- 2025-10-09
  453. vllm setup for nvidia (can use llama) -- 2025-10-05
  454. Full-fine tuning doesn't require much vRAM with gradient checkpointing... -- 2025-10-05
  455. Qwen/Qwen3-Omni-30B-A3B-Thinking -- 2025-10-05
  456. inclusionAI/Ring-mini-linear-2.0 -- 2025-10-05
  457. llama.cpp: Quantizing from bf16 vs f16 -- 2025-10-05
  458. GLM 4.6 is nice -- 2025-10-04
  459. NVFP4 or MXFP4 MOE on sm120 (RTX 5900 RTX 6000 PRO) -- 2025-10-04
  460. Ask Hackaday: How Do You Distro Hop? -- 2025-09-30
  461. Uncensor Qwen3 models without retraining -- 2025-09-20
  462. Depth upscaling? -- 2025-09-20
  463. Definitive proof openai/gpt-oss-20b is dumb as hell -- 2025-09-19
  464. Free 10%+ Speedup for CPU/Hybrid Inference on Intel CPUs with Efficiency Cores -- 2025-09-17
  465. Claude Performance Report with Workarounds - September 7 to September 14 -- 2025-09-16
  466. PSA/RFC: KV Cache quantization forces excess processing onto CPU in llama.cpp -- 2025-09-15
  467. native tool calling support for DeepSeek V3.1 just merged in llama.cpp -- 2025-09-15
  468. model : add grok-2 support by CISC · Pull Request #15539 · ggml-org/llama.cpp -- 2025-09-15
  469. An Afternoon at the Recursive Café: Two Threads Interleaving -- 2025-09-14
  470. blacktop/go-hypervisor -- 2025-09-14
  471. TSYJ-He/AutoEnvForge -- 2025-09-14
  472. Hackaday Links: September 7, 2025 -- 2025-09-14
  473. Any idea how to use ollama (debian) with 2x GPUs to load larger models? -- 2025-09-14
  474. Rails on SQLite: new ways to cause outages -- 2025-09-14
  475. Qwen3-Coder-480B Q2_K_XL same speed as Qwen3-235b-instruct Q3_K_XL WHY? -- 2025-09-09
  476. Renting GPUs is hilariously cheap -- 2025-09-09
  477. Ex-Miner Turned Local LLM Enthusiast, now I have a Dilemma -- 2025-09-09
  478. Tencent-Hunyuan/HunyuanWorld-Voyager -- 2025-09-09
  479. How the “Kim” dump exposed North Korea's credential theft playbook -- 2025-09-09
  480. Further Adventures in Colorimeter Hacking -- 2025-09-09
  481. 🌟Introducing Art-0-8B: Reasoning the way you want it to with Adaptive Thinking🌟 -- 2025-09-02
  482. Fine Tune Model for Home Assistant? -- 2025-09-02
  483. [Editorial] Claude Code - massive issues -- 2025-09-01
  484. TheDrummer is on fire!!! -- 2025-09-01
  485. NVIDIA-Nemotron-Nano-12B-v2 -- 2025-09-01
  486. Gpt-oss Fine-tuning - now with 60K context length and fits on &lt;13GB VRAM -- 2025-08-30
  487. A PLL For Perfect Pitch -- 2025-08-27
  488. Why are users still able to edit system prompts or memories even after disabling it? -- 2025-08-20
  489. Making your prompts better with GEPA-Lite using Ollama! -- 2025-08-15
  490. Optimizing OpenWebUI's speed through indexing (using PostgreSQL as a back-end) -- 2025-08-11
  491. PSA: DuckDuckGo search in OWUI routes to non-privacy friendly providers like Bing, Google, and Yahoo. -- 2025-08-11
  492. Open-webui Tools for Firewalla -- 2025-08-11
  493. Local RAG with 97% smaller index and Claude Code–compatible semantic search -- 2025-08-10
  494. Trump Announces 100% Tariff on Semiconductors, unless made in US -- 2025-08-08
  495. The Tape Speed Keyboard -- 2025-08-08
  496. My first finetune: Gemma 3 4B unslop via GRPO -- 2025-08-02
  497. Supervised Fine Tuning on Curated Data is Reinforcement Learning -- 2025-08-02
  498. Debugging the Pixel 8 kernel via KGDB -- 2025-07-31
  499. Ollama + Open WebUI -- is there a way for the same query to run through the same model multiple times (could be 3 times, could be 100 times), then gather all the answers together to summarise/count? -- 2025-07-25
  500. Localllama’s (first?) IFTA - I’ll Fine-Tune Anything -- 2025-07-20
  501. fsndzomga/metadspy -- 2025-07-10
  502. osmosis-ai/Osmosis-Apply-1.7B -- 2025-07-10
  503. IntervitensInc/pangu-pro-moe-model -- 2025-07-10
  504. Continual Gradient Low-Rank Projection Fine-Tuning for LLMs -- 2025-07-10
  505. Creating custom kernels for the AMD MI300 -- 2025-07-10
  506. i made a commit message generator that can be used offline and for free -- 2025-07-05
  507. THUDM/GLM-4.1V-9B-Thinking -- 2025-07-05
  508. baidu/ERNIE-4.5-VL-424B-A47B-Base-Paddle -- 2025-07-05
  509. LoRA Fine-Tuning Without GPUs: A CPU-Efficient Meta-Generation Framework for LLMs -- 2025-07-05
  510. trycua/cua -- 2025-06-30
  511. ml0-1337/claude-gate -- 2025-06-30
  512. mstrYoda/go-arctest -- 2025-06-30
  513. Accelerating Docker Builds by Halving EC2 Boot Time -- 2025-06-30
  514. Advanced Time Manipulation with GDB -- 2025-06-21
  515. Practical SDR: Getting started with software-defined radio -- 2025-06-21
  516. liaotxcn/Probabilistic-Filters -- 2025-06-19
  517. flohoss/gocron -- 2025-06-19
  518. Lessons from Mixing Rust and Java: Fast, Safe, and Practical -- 2025-06-19
  519. 100 prisoners and a lightbulb -- looking back -- 2025-06-19
  520. Paper2Poster/Paper2Poster -- 2025-06-10
  521. Octoberfest7/zip_smuggling -- 2025-05-30
  522. Silencing Firefox's Chattiness for Web App Testing -- 2025-05-30