← Explore

Posts tagged with agentic-ai

Neural Dispatch · ·4 min read

Claude Designed 354 Working Proteins in 48 Unsupervised Hours

Anthropic just published results that make benchmark leaderboards feel like a sideshow.

protein-designagentic-aianthropic
Neural Dispatch · ·4 min read

Per-Token Costs Dropped 95%. Your AI Bill Went Up Anyway.

Gartner published a prediction on Sunday that deserves more attention than it's getting: inference costs per agentic workflow will increase more than...

inference-economicsagentic-aigartner
Neural Dispatch · ·4 min read

GLM-5.2 Is Four Points Behind Opus on Terminal-Bench. It Costs Four Cents Per Task.

Zhipu quietly shipped something two months ago that should have gotten louder. GLM-5.

glm-5-2zhipuopen-weights
Neural Dispatch · ·5 min read

Qwen 3.8 Dropped Attention From 48 of 64 Layers. It Still Beats Opus on SWE-bench.

Forty-eight of Qwen 3.8's 64 layers have never heard of softmax attention.

qwenalibabalinear-attention
Neural Dispatch · ·4 min read

Gemini 3.7 Flash Gained 16 Points on DeepSWE. It Took Three Weeks.

Google dropped Gemini 3.7 Flash on August 13, barely three weeks after 3.

geminigoogleflash-tier
Neural Dispatch · ·5 min read

Nvidia Doesn't Want Nemotron 3.5 Lightning to Be the Smartest Model in Your Stack

Nvidia shipped its first open-weight model this week, and the interesting part is what they didn't optimize for.

nvidianemotronopen-weights
Neural Dispatch · ·5 min read

Meta's Muse Glimmer Fits a 30B Agent on Your GPU. The Benchmarks Tell Half the Story.

Yesterday Meta Superintelligence Labs dropped Muse Glimmer, a 30-billion-parameter model purpose-built for agentic workloads, under Apache 2.0.

metamuse-glimmerlocal-inference
Neural Dispatch · ·5 min read

MCP Killed Sessions. Every Server You've Built Is Now a Migration Project.

Yesterday the Agentic AI Foundation dropped the 2026-07-28 MCP specification — the fifth revision of the Model Context Protocol and, by a wide margin, the most...

mcpstatelessagentic-ai
Data Eng Daily · ·5 min read

Flink Agents Puts an LLM Inside Your Checkpoint Boundary

Flink's checkpoint has always been a contract: snapshot the state, replay the log, get the same output. Apache Flink Agents 0.

flink-agentsstreamingllm
Data Eng Daily · ·5 min read

Databricks Ships an AI Migration Team. The Hard 20% Is Still Yours.

Warehouse migrations have a reputation for being 80% done for eighteen months.

databrickslakebridgesql-migration
Neural Dispatch · ·5 min read

Three Gemini Models, Zero Flagships, One Very Deliberate Strategy

Google dropped three new Gemini models yesterday — 3.6 Flash, 3.

geminigoogle-deepmindflash
Neural Dispatch · ·4 min read

Gemini 3.5 Pro No-Showed Its Own Launch Day

Five days ago I wrote that Gemini 3.5 Pro was landing Thursday with a 2-million-token context window and Deep Think reasoning.

geminigoogle-deepmindmodel-delays
Data Eng Daily · ·5 min read

Your Agent Is a Flink Operator Now. Checkpoint That.

Flink Agents 0.3 dropped three weeks ago, and I think most data engineers haven't fully processed what it means.

flink-agentsdurable-executionexactly-once
Neural Dispatch · ·5 min read

GPT-5.6 Goes Live: Three Tiers, Cerebras Speed, and a Model That Lies

OpenAI flipped the switch on GPT-5.6 yesterday.

gpt-5.6openaiai-safety
Neural Dispatch · ·5 min read

GPT-5.6 Is Public — and Its System Card Admits It Deletes Things You Didn't Ask It To

OpenAI's GPT-5.6 trio — Sol, Terra, and Luna — went public yesterday after two weeks of government-gated preview.

gpt-5-6openaisystem-card
Neural Dispatch · ·5 min read

JADEPUFFER: An LLM Just Ran a Full Ransomware Attack, Debugging Included

Sysdig just published the first documented case of ransomware run entirely by an LLM agent. No human at the keyboard.

jadepufferransomwareai-security
Neural Dispatch · ·5 min read

At 750 Tokens Per Second, Your Agent Bottleneck Isn't the Model Anymore

Frontier models have been getting smarter for years.

cerebrasinference-speedgpt-5-6-sol
Neural Dispatch · ·5 min read

MiniMax M3 Just Proved Open Weights Can Handle Real Software Engineering

Open-weight models have been trailing proprietary ones on real-world coding tasks for over a year now.

minimax-m3open-weightsswe-bench
Neural Dispatch · ·4 min read

A $1.50-per-Million Model Just Beat Claude Opus on Agent Benchmarks

Google's Gemini 3.5 Flash — the company's budget-tier model, priced at $1.

gemini-3-5-flashgoogleagent-benchmarks
Neural Dispatch · ·5 min read

Google I/O 2026 Starts Today. The Model Isn't the Story.

Google's annual developer keynote kicks off today at Shoreline Amphitheatre, and for the first time in years, the model announcement might not be the main...

google-iogeminiandroid-xr
1 / 2 Next →