Archive

27 published today · 107 in the edition. Everything this publication has kept.

17 Sep 2026Ollama 0.30 adds GGUF support through llama.cpp alongside its MLX engineBrief
17 Sep 2026Ollama updates MLX engine for Apple Silicon with faster, leaner local models claimedBrief
17 Sep 2026Ollama 0.31 speeds up Gemma 4 on Apple Silicon with MLX multi-token predictionBrief
17 Sep 2026Mistral's Robostral Navigate claims single-camera robot navigation at 76.6% on R2R-CEReport
17 Sep 2026Telecom fraud benchmark swaps easy negatives for near-domain calls, and classifier scores fall to 0.65-0.68 Macro-F1Brief
17 Sep 2026A governed pipeline cuts guideline-to-artifact work from days to hours — and lowers hallucinationBrief
17 Sep 2026NeMo Data Designer builds multimodal synthetic datasets from declarative column specs and a preview loopBrief
17 Sep 2026Masking improved compositional generalization in preregistered test of sixty four-cell systemsBrief
17 Sep 2026GraphEcho tests whether graph agents count repeated paths as new evidenceBrief
17 Sep 2026Captions outperform direct video QA on long egocentric video, CapMem benchmark findsBrief
17 Sep 2026GVD links document versions and resolves rule conflicts locally, without an LLMBrief
17 Sep 2026EvolveTrade rewrites a trading agent's system prompt from its own decision traces, leaving the model fixedBrief
17 Sep 2026Mistral Studio adds versioned, traceable system of record for prompts and skillsBrief
17 Sep 2026Inspect Evals v0.14.4 fixes scorer grade-binding and setup behavior in four evaluation suitesBrief
17 Sep 2026METR proposes an 'expenditure horizon' to measure an AI agent's optimization abilityBrief
17 Sep 2026Mistral's Shieldstral puts moderation policy in the prompt, not the model weightsReport
17 Sep 2026Meta's first open model from Superintelligence Labs lands on Ollama: Muse Glimmer, a 30B multimodal build for local coding agentsBrief
17 Sep 2026Anthropic signs deal to lease site of its first Australian datacentre in western QueenslandBrief
17 Sep 2026OpenAI opens a misalignment incident log, with six training-time cases disclosedReport
17 Sep 2026OpenAI sketches an advertising push around Sponsored Agents and HubSpot, Shopify integrationsReport
17 Sep 2026LlamaIndex v0.14.14 adds a token-budget callback and LangChain 1.x supportBrief
17 Sep 2026LlamaIndex 0.14.15 adds multimodal core support and AgentMesh trust layerBrief
17 Sep 2026LlamaIndex 0.14.16 adds built-in rate limiters for LLM and embedding callsBrief
17 Sep 2026n8n Gateway credits let Cloud users call AI models without a provider account or API keyBrief
17 Sep 2026LlamaIndex 0.14.24: parser, ingestion and memory fixes land in coreBrief
17 Sep 2026Mistral's Agentic Search moves retrieval from one-shot chunks to a multi-step loopReport
17 Sep 2026Anthropic merges Claude chat and Cowork, and adds Docs and Slides for document outputReport
16 Sep 2026Sentence Transformers v6.0 adds training for multi-vector, ColBERT-style retrieval modelsReport
16 Sep 2026LangChain puts Managed Deep Agents and its LLM Gateway into public betaBrief
16 Sep 2026Cloudflare gives bot operators a dashboard home with submission tracking, editing and a behavior modelBrief
16 Sep 2026A design-system CLI for coding agents: cheaper tokens, 8% faster tasks, wider tool reachBrief
16 Sep 2026Hugging Face ships a WebGPU kernel loader and 207 versioned kernels for browser inferenceReport
16 Sep 2026BenchMIRT splits a single benchmark score into the capabilities actually driving itReport
16 Sep 2026A coding model taught to paint: open RL recipe puts the reward in a hand-rated reference poolReport
16 Sep 2026inspect_evals v0.14.3: SWE-bench runs that miss a marker now error instead of scoring 0.0Brief
16 Sep 2026funes gives coding agents a memory layer you host yourselfReport
16 Sep 2026NeoMME ships a 260M and 800M multimodal encoder, with a retriever the team says cuts late-interaction index storage 255×Report
16 Sep 2026GPT-6 Astra claims state-of-the-art computer use and coding, with token prices up 2.5xBrief
16 Sep 2026SAGE scores literary quality in six layers, and finds the interpretive gap is roughly double the emotional oneBrief
16 Sep 2026A Benchmark That Scores Markdown Format Failures Separately From ContentBrief
16 Sep 2026SQLMorph mutates text-to-SQL queries to expose join and wording failures, and scores partial answersBrief
16 Sep 2026Mistral describes a 40,000-line Fortran 77 to C++ migration built around a numerical parity harnessReport
16 Sep 2026Cloudera and Mistral partnership targets on-prem and air-gapped model deploymentReport
16 Sep 2026Atlassian says it is bringing governed agent loops to its AI-native SDLCBrief
16 Sep 2026AWS pairs agent quality scoring with autonomous infrastructure investigation for production multi-agent systemsBrief
16 Sep 2026Perplexity runs GPT-6 Astra across systems work and checks in less oftenReport
16 Sep 2026Fyxer's AI assistant leans on fine-tuning, memory and user feedback to sound like its ownerReport
16 Sep 2026AWS post outlines unattended replenishment loop with Databricks Genie and Amazon QuickBrief
16 Sep 2026Amazon Bedrock AgentCore adds a consent portal for end-user OAuth with GitHub and Slack targetsBrief
16 Sep 2026Abnormal AI runs Bedrock AgentCore Code Interpreter as an agent scratch pad for email threat detectionBrief
16 Sep 2026LangChain guide collects production agent lessons from Schneider Electric, Vodafone and monday.comBrief
16 Sep 2026Claude Fable 5.1 returns to Copilot after upstream provider outageBrief
16 Sep 2026Copilot budget increase requests are now generally availableBrief
16 Sep 2026An agent that passed 77.4% of runs passed all five on only 53% of tasks. IBM's new diagnostic targets that gap.Report
16 Sep 2026llama.cpp b10996 forces a newline at the end of the reasoning budget for qwen3-coder in chatBrief
16 Sep 2026llama.cpp b10997 extends MoE tile heuristic to RDNA3.5 on ROCmBrief
16 Sep 2026Mozilla's Firefox Smart Window beta runs on Mistral models, starting in France and North AmericaReport
16 Sep 2026GitHub AI Scan for pull requests no longer requires CodeQL default setupBrief
16 Sep 2026llama.cpp b11009 fixes tensor-parallel split states for fused QKV in gemma4 and qwen35Brief
16 Sep 2026TypeSafe AI releases Jev, a model that returns classifications instead of textBrief
16 Sep 2026AgentCore's system prompt optimizer proposes changes from production traces, then validates them before promotionBrief
16 Sep 2026BearQ brings an assignable testing agent into Jira workflowsBrief
16 Sep 2026smartARM's bionic arm selects grips automatically using Meta's open-source DINOv2 vision modelBrief
16 Sep 2026Spain's AEPD receives first report of an AI-agent data breachReport
16 Sep 2026Benchmark Radar indexes 1,283 benchmark records and ships a CLI for prior-art searchesBrief
16 Sep 2026Attackers are routing malware through shared AI artifacts, conversations and search adsBrief
16 Sep 2026vLLM v0.29.0 makes Model Runner V2 the default and adds per-request speculative decoding metricsBrief
16 Sep 2026AWS publishes 38 open-source agent skills aimed at HCLS reasoning errorsBrief
16 Sep 2026Security guide flags exposed API keys as a common way vibe-coded apps failBrief
16 Sep 2026New reporting framework separates implementation from evaluation in AI-assisted systematic reviewsBrief
16 Sep 2026Zapier highlights AI model flexibility to avoid lock-inBrief
16 Sep 2026AI agents add decision-making to business automationBrief
16 Sep 2026Open-source harness measures cost per correct answer for OpenAI models on Amazon BedrockBrief
16 Sep 2026Shopify CLI 4.8 adds terminal commands for dev store managementBrief
16 Sep 2026llama.cpp fixes Metal NaN for activations beyond f16 rangeBrief
16 Sep 2026Small correction module fixes some language model errors without degrading base skillsBrief
15 Sep 2026Claude Code adds forking of remote-control sessions as background jobsBrief
15 Sep 2026AI Trained on Railroad Game Improved at Financial Research, But Only With the Right Training DesignReport
15 Sep 2026China's Spy Chief Flags US AI Models for Faster Cyber Vulnerability DiscoveryBrief
15 Sep 2026Mistral 7B: a 7B model that outperforms Llama 2 13B, released under Apache 2.0Report
15 Sep 2026Llama 3.2 comes to Ollama with small local text models and vision models on the wayReport
15 Sep 2026OFASys Enables Multitask Learning with One Line of CodeBrief
15 Sep 2026Bespoke-Minicheck, a new factuality checking model, arrives in OllamaBrief
15 Sep 2026Guide Introduces 'Critique Shadowing' for LLM EvaluationBrief
15 Sep 2026Llama 3.2 Vision Runs Locally in Ollama, Adding Image Analysis to Your MachineReport
15 Sep 2026Minions: a method for local and cloud LLM collaborationBrief
15 Sep 2026AI Evals FAQ Draws a Line Between Model Benchmarks and Product EvalsBrief
15 Sep 2026Qwen3-Coder: New Agentic Code Model with 1M Token ContextBrief
15 Sep 2026Qwen-Image-Edit adds precise text editing to image editingBrief
15 Sep 2026Ollama's new scheduler targets out-of-memory crashes and GPU utilizationBrief
15 Sep 2026MiniMax M2 arrives on Ollama’s cloud for coding and agentic workflowsReport
15 Sep 2026Gemini 3 Flash promises frontier performance at lower cost and global accessReport
15 Sep 2026Three Years from GPT-3 to Gemini 3: The Move from Chatbots to AgentsReport
15 Sep 2026Gemini app adds Lyria 3 music generation for 30-second tracksReport
15 Sep 2026Gemini 3.1 Flash TTS Adds Granular Audio Tags for Precise Speech ControlReport
15 Sep 2026Claude Code v2.1.267 Adds Effort Cap and Fresh System Prompt OptionBrief
15 Sep 2026Claude Code v2.1.268 adds gateway pricing setting for consistent cost trackingBrief
15 Sep 2026Claude Code v2.1.271 adds per-command domain allowlists for sandboxed commandsBrief
15 Sep 2026OdoBot cuts web-agent token use by 44-80% on a 45-task testBrief
15 Sep 2026Judge feedback narrows the gap between cheap and costly patent-drafting modelsBrief
14 Sep 2026GitHub Copilot auto model selection adds efficiency, balance, and intelligence tiersBrief
14 Sep 2026Competence gate improves hybrid forecasting by weighting language models per domainBrief
12 Sep 2026Autonomous AI Agents: What They Can Do and How to Manage the RisksReport
12 Sep 2026OpenRouter's GPT Sol Latest Alias Always Points to the Newest GPT Sol ModelReport
12 Sep 2026Agent can study an unfamiliar environment before test time, paper findsBrief
12 Sep 2026Nemotron specialists reach IMO gold threshold with open natural-language pipelineBrief
12 Sep 2026Program-solve interface improves clinical math for 32B model, not 7BBrief