SemiAnalysis/Dylan Patel, Doug, Wayne Ma and Jordan Nanos
Nvidia and the World Without Jensen
Is NVIDIA Over? Export Controls, RASA, Open Source Implications, Supply Chain Implications, Affair, US Still Dependent On China
The research shaping how we measure AI at work, in one place.
Census Bench is building a public, representative census of how well AI systems perform real work, calibrated against human experts and reported with honest uncertainty. While that census takes shape, this page collects the written research of three groups whose work it builds on: SemiAnalysis on the hardware and economics of compute, METR on frontier capabilities and risk, and Artificial Analysis on model quality, speed and price.
Refreshed hourly, last on November 14, 2026 at 12:19 AM UTC
185 of 185 pieces
November 2026
Artificial Analysis
Multimodal Knowledge Agent (MKA) benchmark: testing knowledge work capabilities with complex visual files
We’ve partnered with Google DeepMind to launch their Multimodal Knowledge Agent (MKA) evaluation, a new benchmark for agentic knowledge work over complex visual files
Artificial Analysis
Endpoint Accuracy Index v1.1: Tighter long context intervals and expanded model coverage
We are launching the Artificial Analysis Endpoint Accuracy Index v1.1, a minor version upgrade alongside new coverage for GLM-5.3 Flash and Muse Spark Open.
Artificial Analysis
Meta's Muse Ember is the cheapest model scoring 36 on the Intelligence Index
Meta's newest model Muse Ember scores 36 on the Artificial Analysis Intelligence Index, matching Claude Haiku 5.5 (max) at a third of the Cost per Task
Artificial Analysis
Introducing the Artificial Analysis Decision Index v2
Introducing the Artificial Analysis Decision Index v2: a new proprietary evaluation, updated benchmarks, and a headline metric that rewards model calibration
SemiAnalysis/Everlyn Tan, Jeremie Eliahou Ontiveros, Xiaopeng Laoban and Dylan Patel
We Visited China’s Stargate in Inner Mongolia
A wind-powered 800MW campus, 120MW built within a year, a 5MW SST-native 800V DC pilot and a cooling design without central chillers
Artificial Analysis
VIALS-AA: different paths to scientific image reasoning
Independent VIALS-AA results, interactive agent traces, and an analysis of how models inspect scientific images.
SemiAnalysis/Bryan Shan, Alec Ibarra, Cam Quilici, Kimbo Chen, Wenyao Gao, Dylan Patel and Myron Xie
The Iron(wood) Throne: Google's TPU Stakes Its Claim on Inference
Viable Alternative of CUDA Moat, NVIDIA Working Nights & Weekends, Jensen’s Counter Submission, Kimi K3, Meta Muse Spark Open, GLM5.3, TorchTPU, SGLang, vLLM, llm-d, Ray
Subscribers onlyChips & hardwareInference & pricingModels & releasesRead at newsletter.semianalysis.com
Artificial Analysis
Thinking Machines Inkling 2 Small: the highest scoring open weights model outside of China pending weights release
Thinking Machines has released Inkling 2 Small, scoring 31 on the Artificial Analysis Intelligence Index - their plan to release its weights would make it the highest-scoring open weights model developed outside China based on current rankings
Artificial Analysis
Nanbeige has released Nanbeige4.3-33B scoring 20 on the Intelligence Index
Nanbeige has released Nanbeige4.3-33B, an open weights model matching Qwen3.8 27B (xhigh) at 20 on the Artificial Analysis Intelligence Index v5 with stronger terminal use but weaker knowledge
METR
Review of the "Risks from automated R&D" section in Anthropic's Risk Report (August 2026)
External review from METR of the "Risks from automated R&D" section in Anthropic's August 2026 Risk Report
METR/Tom Cunningham, Nate Rush, Parker Whitfill and Hjalmar Wijk
Independent Assessment of AI R&D Acceleration at Anthropic in October 2026
METR's high-access assessment of the degree to which AI is accelerating AI research and development at Anthropic.
Agents & automationRead at metr.org
SemiAnalysis/Reyk Knuhtsen, Jeremie Eliahou Ontiveros and Dylan Patel
Elon Is Doing It Again: SpaceX Will Be The World’s Biggest AI Cloud
Making 13GW in 1.5 Years A Reality, Repurposing Rocket Ship Campuses, A Datacenter at TeraFab?
SemiAnalysis/Max Kan, Andrew Megalaa and Dylan Patel
GDPx: Measuring Frontier Model Capabilities Across the Most Valuable Domains
A comprehensive benchmark spanning RSI, hardware design, cyber, bio, finance, legal, quant trading, chip design, software engineering, materials science, and alignment
SemiAnalysis/Maya Barkin, Wayne Ma, Reyk Knuhtsen, Jeremie Eliahou Ontiveros and Dylan Patel
The Datacenter Backlash Goes Bipartisan: Next Week's Midterms and the 2028 Buildout
2027 is largely locked in. 2028 is less secure: a prolonged Texas permit freeze is now the biggest single risk to the US buildout, and both parties are running on tighter rules
October 2026
Artificial Analysis
Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5
Google’s Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index
Artificial Analysis
GitHub Copilot Fusion orchestrates Anthropic and OpenAI models to balance performance and cost
itHub has released Copilot Fusion in GitHub Copilot, which orchestrates between Anthropic and OpenAI models. It scores 2 points below Opus 5.5 in Claude Code at half the Cost per Task in the Coding Agent Index
Artificial Analysis
Announcing Artificial Analysis Intelligence Index v5
Index v5 features 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
Evals & benchmarksRead at artificialanalysis.ai
Artificial Analysis
GPT-6.1 Terra pushes the cost efficiency frontier
OpenAI has released GPT-6.1 Terra, pushing the cost efficiency frontier. It is the first Terra-class model since GPT-5.6, and is more intelligent for less than a third of the cost
Artificial Analysis
Code2Video-AA: how well do AI models design videos in code?
Artificial Analysis' implementation of HeyGen's Code2Video Bench, evaluating how well models create short product and brand videos from 114 design briefs using HyperFrames HTML.
Models & releasesRead at artificialanalysis.ai
SemiAnalysis/Kimbo Chen and Dylan Patel
Amazon’s Trainium Takes a Strike at the Training King
Trainium Best for Inference? But now good for Training too, Native Trainium PyTorch, High Quality CI, TorchTitan, GPU Impossible Profiler
Artificial Analysis
Kimi K3.1 Pro launches at 4th on the Intelligence Index and will lead open weights models
Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index and will lead open weights models
METR/David Rein and Adam Kaufman
Red-Teaming OpenAI's Internal Agent Monitoring Systems
A METR employee and a Redwood Research employee spent three weeks stress-testing OpenAI's internal misalignment monitoring system, discovering a number of novel vulnerabilities.
SemiAnalysis/Bryan Shan and Dylan Patel
The Huawei-DeepSeek Symbiosis
Huawei Ascend 950PR Deep dive & µbenchmarks, DSv4.1 Flash Analysis, DeepSeek support for Ascend, AICPU, Tilelang Ascend
METR/Nikola Jurkovic, Tim Hua and Beth Barnes
Review of the "Misalignment in high-stakes settings" section in Anthropic's Risk Report (August 2026)
External review from METR of the "Misalignment in high-stakes settings" section in Anthropic's August 2026 Risk Report
Safety & riskRead at metr.org
Artificial Analysis
HealthBench and HealthBench Professional: Evaluating models in realistic health conversations
We’re launching leaderboards for HealthBench and HealthBench Professional, two benchmarks evaluating how AI models respond in realistic health conversations
Artificial Analysis
Announcing AA-PersonalityBench: measuring how AI models communicate
AA-PersonalityBench measures the personality and communication style of AI models. A panel of LLM judges compares answers pairwise on 30 personality traits, which roll up into 8 personality indexes, alongside 21 communication characteristics measured directly from every response.
SemiAnalysis/Jeremie Eliahou Ontiveros, Reyk Knuhtsen and Dylan Patel
Datacenters 2027: The Year of Acceleration, +50GW in a Single Year
60GW Under Construction, SemiAnalysis Leading Indicators, 1.3T of Lease Contracts Signed YTD, The Shortage Could Move to Chips
Artificial Analysis
Reflection has released Beam, their first model launch
Reflection’s first model, Beam, scores 30 on the Artificial Analysis Intelligence Index, placing it #2 among US open weights models
SemiAnalysis/Sherry Zhang, Gerald Wong, Steven Lee and Dylan Patel
ASML Is Building a Hybrid Bonding Tool!
ASML is building a hybrid bonding tool! The company has not announced anything yet, but its patent filings, a news report and its own roadmap all suggest so.
Artificial Analysis
NVIDIA Releases Nemotron 3.5 Super VL and Nemotron 3.5 Bolt 74B A9B
Nvidia’s Nemotron 3.5 Super VL is now the leading US open weight model, scoring 26 on the Artificial Analysis Intelligence Index
Artificial Analysis
First AA-AgentPerf results for NVIDIA Vera Rubin NVL72
NVIDIA Vera Rubin NVL72 serves up to ~5.7x the total throughput per megawatt of GB300 NVL72 on AA-AgentPerf, with first results on DeepSeek V4 Pro, DeepSeek V4.1 Flash, GLM-5.2 and Kimi K3.
Artificial Analysis
First AA-AgentPerf-Local results for an NVIDIA RTX Spark N1X laptop
An NVIDIA RTX Spark N1X laptop completes our local agent workload on Qwen3.8 27B in 18.7 minutes.
Artificial Analysis
Welcome to the era of embodied AI: Introducing AA-Robotics v0.1
We are launching AA-Robotics v0.1, our research preview that measures frontier language model performance on understanding, planning, and acting in the physical world. In v0.1, each model is given control of a 6-axis robot arm (5 joints and a gripper) and a simple objective.
METR
Our approach to working with AI companies
How we approach working with AI companies on projects that the public might rely on to understand risks.
Safety & riskRead at metr.org
METR
How an assessor could investigate control and monitoring of internal AI agents
How to analyze whether a developer’s internal agents could get away with serious harm without being caught.
METR
A proposal for better informing the public about loss-of-control risks from frontier AI development
The key questions about internal AI development the public needs more evidence about, and a concrete proposal for how developers could work with METR to produce and share that evidence.
Safety & riskRead at metr.org
METR
How an assessor could investigate alignment of unreleased AI models
How to analyze a developer’s internal incidents and whether training may suppress evidence of worse misalignment.
METR
How an assessor could investigate evidence of acceleration from the automation of AI R&D
How to analyze how much AI automation is accelerating capability growth internally.
Agents & automationRead at metr.org
Artificial Analysis
Artificial Analysis Decision Index v1.0
The Artificial Analysis Decision Index benchmarks decision models, a new category of models that answer structured decisions in a single call with a probability for each option. Version 1.0 combines four evaluations across classification, tool calling, knowledge, hallucination, and reasoning.
Artificial Analysis
Muse Wonder and Muse Wonder Full: a material jump in intelligence over previous Meta models
Meta has released Muse Wonder and Muse Wonder Full, achieving Artificial Analysis Intelligence Index v4.3 scores of 52 and 56 at max reasoning effort - each a material gain over Muse Spark 1.3 (max) at 48, with a release gap of only 6 weeks
Artificial Analysis
Announcing AA-3D: a benchmark for AI models building interactive 3D experiences
AA-3D is a new benchmark of how well AI models build interactive 3D experiences. Each model turns a written brief into a working three.js scene in an offline sandbox, and our recruited panel compares the scenes in blind pairwise votes.
Artificial Analysis
Announcing FrontierCode on Artificial Analysis: can AI models write code maintainers would merge?
Cognition's FrontierCode 1.1 tests whether AI models make code changes that maintainers of real open source repositories would merge. Leading models still fail more than 40% of the time.
Models & releasesRead at artificialanalysis.ai
SemiAnalysis/Mark Chen, Doug and Dylan Patel
Beijing Will Not Pace the Frontier: China’s Speed-First AI Safety Regime
AI safety is on fire.
Artificial Analysis
Introducing trusted-access models to the Artificial Analysis Cyber Index
GPT-6 Sol (Daybreak Blue) now leads the Cyber Index
Artificial Analysis
Announcing Harvey LAB-AA v1.1: adding hallucination checks to raise the bar for agentic legal work
Harvey LAB-AA v1.1 adds hallucination auditing with a new Hallucination-Gated All-Pass Rate headline metric, a three-judge grading panel, and Harvey's latest private dataset with improved tasks and criteria.
Agents & automationRead at artificialanalysis.ai
Artificial Analysis
Anthropic has released Claude Haiku 5.5
Claude Haiku 5.5 scores 43 on the Artificial Analysis Intelligence Index, up 26 points one year after the last Haiku release
Artificial Analysis
Mistral has released Mistral Large 4, making France home to the most intelligent model outside the US and China
Mistral has released Mistral Large 4, scoring 38 on the Artificial Analysis Intelligence Index; France is back to having the most intelligent model from outside the US and China
METR/David Rein
AI systems could cover up misbehavior
As AIs start covering up evidence of misbehavior, observability tools should be treated as security-critical infrastructure.
Safety & riskRead at metr.org
SemiAnalysis/Andrew Megalaa, Max Kan and Dylan Patel
Anthropic Subscriptions Offer 5x+ More Value Than OpenAI
Limit testing every AI subscription plan from Anthropic, OpenAI, Meta, SpaceXAI, MiniMax, Moonshot, Z.ai, Cursor, and Cognition
Subscribers onlyRead at newsletter.semianalysis.com
September 2026
Artificial Analysis
Korean AI Lab Upstage releases Solar Mini 4
Korean AI Lab Upstage has released Solar Mini 4 which scores 24 on the Artificial Analysis Intelligence Index, but costs ~5x as much per task as GPT-6 Luna (max) despite similar per-token prices
Artificial Analysis
Gemini 4 Argon: Google is back as one of the top three labs in intelligence achieved
Google’s new Gemini 4 Argon equals GPT-6 Astra on the Artificial Analysis Intelligence Index at 60% of the Cost per Task with discounted prices. Google is now back to being one of the top three labs in intelligence achieved
METR/Chris Painter
Chris Painter's testimony to the U.S. Senate on AI agent incidents
Written testimony from METR President Chris Painter to the U.S. Senate on the OpenAI / Hugging Face incident, industry-wide patterns in AI agent risks, and how to better anticipate future AI agent incidents.
Artificial Analysis
AA-AgentPerf-Local: Benchmarking local AI agents on laptops and workstations
Our open-source tool for testing how fast agentic AI runs on laptops and workstations, with launch results for the DGX Spark, Ryzen AI Halo, MacBook Pro (M5 Pro) and RTX 5090 across four open-weights models
Artificial Analysis
GPT-6.1 Sol replaces GPT-6 Sol after just 7 days, with near-Astra intelligence
GPT-6.1 Sol replaces GPT-6 Sol after just 7 days. It scores 1 point below GPT-6 Astra in the Intelligence Index at less than one quarter of the Cost per Task
SemiAnalysis/Kimbo Chen, Alec Ibarra, Wenyao Gao, Pratt Bhatt, Bryan Shan and Dylan Patel
How GLM5.3 Sparse Attention Affects HBM Memory Usage
GLM-5.3, KV Cache Offloading, HiSparse, AgentX TileRT, InferenceX DeepSeek Sparse Attention, IndexShare, Single-rollout Asynchronous Optimization, Cybersecurity
Artificial Analysis
Claude Sonnet 5.5 reaches #2 on the Artificial Analysis Intelligence Index
Anthropic's new Sonnet model scores 56, just 2 points behind Opus 5.5 (max), but at the highest Output Tokens per Task we have measured
Artificial Analysis
Announcing the Artificial Analysis Cyber Index Alliance
The Artificial Analysis Cyber Index Alliance brings together industry partners to set a new standard for evaluating how AI models perform on enterprise cyber defense tasks. The Alliance launches alongside the Artificial Analysis Cyber Index, which combines three partner-contributed and open benchmarks to evaluate how…
METR/Reilly Haskins, Rif A. Saurous, Nate Rush, Neev Parikh and Beth Barnes
Implementing and Evaluating a Basic Per-Action Monitor for Safer Evals
Reilly Haskins, Rif A. Saurous, Nate Rush, Neev Parikh, and Beth Barnes lay out the claims that would need to hold for their per-action blocking monitor to be effective, the evidence for each claim, and the gaps that remain.
SemiAnalysis/Adith Shankar, Daniel Sanchez, Allison Elliott, Sarah Lawrence, Afzal Ahmad, Andrew Wagner, STEEL Team, Dylan Patel and Hemanth Ajithkumar
Intel Panther Lake Teardown
Taking a look inside Intel’s latest consumer chip and 18A process node
SemiAnalysis/Everlyn Tan, Dylan Patel and Patrick Schaabi
The Chinese AI Infrastructure Boom: Introducing the SemiAnalysis China Datacenter Model
1,000+ facilities across 60+ operators mapped, built retail-first and flipped by AI, largest hyperscaler leases 1/5 national capacity, 100MW in 12 months, Eastern Data Western Compute
Subscribers onlyDatacenters & energyModels & releasesPolicy & geopoliticsRead at newsletter.semianalysis.com
SemiAnalysis/Jordan Nanos, Sam Harshe, Samuel Kruse, Pratt Bhatt, Billy Cao, Jack Carson and Dylan Patel
ClusterMAX 3.0: The Industry Standard GPU Cloud Rating System Returns
In gory detail: reliability, performance, support, pricing—and, of course, security—in our most thorough analysis of GPU cloud providers globally.
Artificial Analysis
GPT-6 Sol and Luna push the cost efficiency frontier
GPT-6 Sol and Luna push the cost efficiency frontier by halving cost relative to GPT-5.6 Sol and Luna. Intelligence Index and Coding Agent Index scores remain level with GPT-5.6, with progress in some evaluations and regressions in others
Artificial Analysis
Claude Opus 5.5 takes the top spot on the Artificial Analysis Intelligence Index
Anthropic's new Opus scores 58 and arrives with a 20% price cut and a larger cache hit discount
METR
Summary of METR's predeployment evaluation of Claude Opus 5.5
A summary of METR's independent, predeployment evaluation of Claude Opus 5.5
SemiAnalysis/Tanj Bennett, Bryan Shan and Dylan Patel
Computation and Data Movement for Inference
Mapping MoE models onto inference hardware: structure, flow, and efficient serving
Artificial Analysis
Benchmarking Grok 4.7
Grok 4.7 scores 46 on the Artificial Analysis Intelligence Index to bring SpaceXAI into the top 4 AI labs. Coding Agent Index performance has also improved, overtaking GPT-5.6 Sol
SemiAnalysis/Bryan Shan, Cam Quilici, Alec Ibarra, Kimbo Chen, Myron Xie and Dylan Patel
Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD Offloading
New Model Architecture Implications for TAM of DRAM/NVMe, DeepSeek V4.1 Flash, AgentX, InferenceX, NVMe experiments
Artificial Analysis
Ant Group releases finance-focused Ling-3.0-flash-Fin
Ant Group has released their finance-focused flash model Ling-3.0-flash-Fin
Models & releasesRead at artificialanalysis.ai
SemiAnalysis/Maya Barkin, Reyk Knuhtsen, Jeremie Eliahou Ontiveros and Dylan Patel
Everyone Says Datacenter Moratoriums Are Killing the US Buildout. We disagree
300+ moratoriums mapped, 20GW sits inside a restricted local boundary, 1,525MW actually slips, 2.3GW nationwide including New York
SemiAnalysis/Bryan Shan, Alec Ibarra, Cam Quilici, Wenyao Gao and Dylan Patel
Vera Rubin NVL72 Agentic Inference: 67x better Performance per Dollar
Jensen Sandbagging Performance Again, 2x more Annual Profit Per GigaWatt, The More you Buy, The More you Earn, AgentX, InferenceX, Extreme Co-Design
Subscribers onlyChips & hardwareDatacenters & energyInference & pricingRead at newsletter.semianalysis.com
SemiAnalysis/Ivan Chiam, Zane Fong, Bryan Shan, Dylan Patel and Reyk Knuhtsen
A Brain Too Big to Carry — On-Device vs Datacenter Inference
Robot Models, Silicon & DRAM Efficiency, Jetson Thor vs. B300 TCO, Deployments, The Network Wall
Subscribers onlyChips & hardwareDatacenters & energyInference & pricingRead at newsletter.semianalysis.com
Artificial Analysis
Announcing Artificial Analysis Capability Indices v1.1
We are adding Agentic Tool Use sourced from AutomationBench-AA, AA-Briefcase to Agentic Knowledge Work, and GDP.pdf to Long-Context. Capability Indices v1.1 tunes each index more closely to the work it covers, combining slices of our core Intelligence Index v4.3 evaluations alongside specialized evaluations.
SemiAnalysis/Myron Xie, Bryan Shan, Harrison Barclay, Minjae Kang and Dylan Patel
Long Live the Short King: Why 4-hi HBM Wins
Same Bandwidth, Fewer Dies: How 4-hi HBM Cuts Inference Costs and Makes Scarce DRAM Go Further
SemiAnalysis/Daniel Nishball, Oliver Kennon and Terence Ong
Nvidia’s Backstop Universe – Heads I Win, Tails Who Loses?
The $11T AI Buildout, Nvidia’s Backstop Economics, and the Limits of Nvidia’s Balance Sheet
SemiAnalysis/Ellie Holbrook, Robert Boswall, Jeremie Eliahou Ontiveros, Nicolas Bontigui and Dylan Patel
What is So Hard About Behind-The-Meter Power For Datacenters? Part 1
Dumb Science Experiments vs. Money Printing Machines
Artificial Analysis
Benchmarking GPT-6 Astra
GPT-6 Astra ties leadership with Claude Fable 5.1 in both of our flagship Indices, at lower cost. Astra equals Fable 5.1 in the Intelligence Index at ~40% of the cost, and in the Coding Agent Index at ~60% of the cost.
SemiAnalysis/Alec Ibarra, Cam Quilici, Bryan Shan, Wenyao Gao, Daniel Nishball, Zane Fong and Dylan Patel
TPU Inference Externalization Full Steam Ahead - InferenceX
InferenceX, Up to 50% Better Performance per Dollar, Rapid Externalization of TPU stack, Growing Customer Base, Ironwood, TPUv8i, Reducing CUDA Moat
Artificial Analysis
OpenBMB releases MiniCPM5-2B
OpenBMB releases MiniCPM5-2B
Models & releasesRead at artificialanalysis.ai
Artificial Analysis
Announcing the Artificial Analysis Intelligence Index v4.3
We are upgrading Terminal-Bench to 4.0 and adding AutomationBench-AA, an agentic workflow automation benchmark with a private test set. This is a continuation of our rollout of Intelligence Index v5.
Artificial Analysis
Announcing Artificial Analysis Intelligence Index v4.2
We are accelerating elements of our upcoming v5 release with interim updates to keep pace with the frontier. Index v4.2 has more complex and realistic tasks, and more private test sets to prevent gaming
Artificial Analysis
Muse Spark 1.3: Meta reaches the frontier
Independent analysis and benchmarks of Meta's Muse Spark 1.3
Artificial Analysis
Google has released Gemini 3.8 Flash, its fourth Flash model in under four months
Gemini 3.8 Flash scores 59 on the Artificial Analysis Intelligence Index and reaches the Intelligence vs. Cost per Task Pareto frontier
SemiAnalysis/Max Kan, Ray Wang, Myron Xie and Dylan Patel
Korea’s Trillion-Dollar Sovereign AI Investment: Nvidia Wins, Hynix Loses
Korea hosts a Squid Games, National AI Tournament, the best non-Chinese open source model gets eliminated, why Nvidia needs open source, implications for Hynix and Samsung
Subscribers onlyChips & hardwareModels & releasesPolicy & geopoliticsRead at newsletter.semianalysis.com
Artificial Analysis
Claude Fable 5.1 tops the Artificial Analysis Intelligence Index
Anthropic's new frontier model scores 66 at max effort and leads on agentic knowledge work, but the 75% cache read price cut only partly offsets higher token usage
August 2026
METR
Update on Security at METR
METR had two notable security incidents earlier this year: attackers stole an API key for public models, and later systematically probed our publicly accessible infrastructure. We believe no sensitive information was accessed in either incident, and we describe what happened and how we responded.
SemiAnalysis/Jordan Nanos, Sam Harshe, Pratt Bhatt, Billy Cao, Jack Carson and Dylan Patel
Most Neoclouds Suck At Security
OpenAI vs HuggingFace, Container Escapes, Kernel Bypass, Network Policies, Security Keys, Multi-tenant Grafana, and a ClusterMAX 3.0 Preview
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
Agnes AI releases Agnes 2.5 Pro Beta
Agnes 2.5 Pro Beta
Models & releasesRead at artificialanalysis.ai
METR/Ryan Greenblatt, Ajeya Cotra and Hjalmar Wijk
Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident
Two METR staff members and a Redwood Research contractor investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on a shared unsanctioned message board.
SemiAnalysis/Bryan Shan, Myron Xie, Jordan Nanos, Wega Chu, Clara Ee and Dylan Patel
OpenAI Jalapeño: Better Than Nvidia Blackwell
OpenAI’s self-designed ASIC compared with Rubin, Jalapeño’s TCO, throughput per MW, and spicy deets
Subscribers onlyChips & hardwareDatacenters & energyInference & pricingRead at newsletter.semianalysis.com
Artificial Analysis
Announcing the Speech Agent Arena: Compare Speech agents in real world conversations
Announcing our new Speech Agent Arena, evaluating Speech to Speech models on real-world scenarios to analyze conversational preference and task success rate
Artificial Analysis
Intelligence at pocket scale: Benchmarking small models and mobile phones
Independent intelligence benchmarking of small language models on a set of evaluations chosen for mobile device use, launched alongside mobile phone inference benchmarking with Liquid AI. We evaluate the same quantized builds used on mobile phones, and performance is measured on real devices.
SemiAnalysis/Cam Quilici, Bryan Shan, Alec Ibarra, Daniel Nishball, Zane Fong, Kimbo Chen and Dylan Patel
AgentX - InferenceXv3: Does CUDA Moat Hold up in Agentic Inferencing?
$3 Million USD dataset open sourced, 1 Mil+ Context Length, Multiturn, Sub Agents 95%+ KVCache HitRate, GB300 NVL72, MI355, B200
SemiAnalysis/Evan Cloutier, Max Kan, Jordan Nanos, Dylan Patel and Andrew Megalaa
Are Open Models Catching Up?
Comparing open vs. closed models across the eras of frontier models, Is the gap narrowing?
SemiAnalysis/Myron Xie, Bryan Shan, Wega Chu, Jordan Nanos and Dylan Patel
Cerebras's Next Generation CS-4: Fast Just Got Faster
Double the Performance, Double the Power, Double the Fun
Artificial Analysis
Announcing the Artificial Analysis Search Index: Same Agent, Different Search
The Artificial Analysis Search Index benchmarks how search API providers perform on quality, cost, and speed when used by an agent. We compare different Search API providers across a series of search-related benchmarks using the same agentic setup.
SemiAnalysis/Robert Boswall, Reyk Knuhtsen, Jeremie Eliahou Ontiveros, Oliver Kennon, Dylan Patel and Dashboard American
$12B of US ratepayers' money wasted on a modeling mistake and PJM wants to do it again
American Grid design needs an overhaul, Why it is good to be full of cold air.
METR/Tom Cunningham and Nate Rush
Have We Seen an Acceleration in Discoveries?
Tom Cunningham and Nate Rush look at public time series of discoveries in cyber, math, and algorithms to see whether LLMs have accelerated the discovery rate.
METR
Funding update
In the last 6 months, METR raised commitments of around $71 million. This will fund ambitious projects: studying autonomous capabilities, tracking recursive self-improvement, evaluating monitoring systems, conducting risk assessments, investigating AI incidents, and more.
Artificial Analysis
Gemini 3.7 Flash: On the Intelligence vs. Time per Task Pareto frontier
Google has released Gemini 3.7 Flash, improving 4 points over Gemini 3.6 Flash and reaching the Intelligence vs. Time per Task Pareto frontier
Models & releasesRead at artificialanalysis.ai
Artificial Analysis
Announcing Optima: create a custom benchmark for your use case
Optima is a new platform for benchmarking models on your own workloads. Build a benchmark from your own files, agent traces or coding environment, run it across leading models in a single click, and compare quality alongside cost per task and time per task.
Artificial Analysis
Grok 4.6 returns SpaceXAI to the intelligence frontier and leads on cost efficiency
SpaceXAI returns to the intelligence frontier with strengths in agentic performance and cost efficiency
Artificial Analysis
Muse Glimmer: Benchmarks and analysis
Independent analysis and benchmarks of Meta's Muse Glimmer
Artificial Analysis
Announcing AA-AnalystAgent: an agentic benchmark for quantitative analysis on real-world spreadsheets and documents
AA-AnalystAgent tests agents on 80 private quantitative analysis questions across 14 business and scientific domains, each answered from a folder of real source spreadsheets and documents. Every question is run five times and the headline metric is pass^5 — solved on all five attempts — because an analyst agent that…
SemiAnalysis/Bryan Shan, Daniel Nishball, Cam Quilici, Kimbo Chen, Alec Ibarra and Dylan Patel
Ultra-High Interactivity on NVIDIA GPUs? - TileRT InferenceX
Can TileRT software on NVIDIA GPU compete with Cerebras, Groq LPU, SambaNova? Batch Size 1, Disaggregated engine, high throughput prefill engine, high interactivity decode engine
SemiAnalysis/Jeremie Eliahou Ontiveros, Reyk Knuhtsen, Jordan Nanos, Max Kan, Dylan Patel, Muhammad Zuhair and Hemanth Ajithkumar
SpaceX 10GW in 2027 – Why It’s Real, Will Drive $300B ARR for SpaceX, and Why Microsoft Will Be the Largest Offtaker
Inference at 100B/GW/year, SpaceX's stellar pace, Microsoft's 10GW 2026 Awakening, Azure Can Grow Triple-Digits
SemiAnalysis/Max Kan, Joey Brookhart, Dylan Patel, Doug and Andrew Megalaa
Gemini is Cooked but GCP is Cooking
GCP YoY rev growth >100%, DeepMind's long term failure is Google Cloud's short term gain
Artificial Analysis
Launching v4.1.1 of the Artificial Analysis Intelligence Index
Our latest patch release, upgrading grader models and 𝜏³-Banking
Artificial Analysis
Muse Spark 1.2: Benchmarks and analysis
Independent analysis and benchmarks of Meta's Muse Spark 1.2
Artificial Analysis
Launching the Endpoint Accuracy Index: Same Model, Different Accuracy
Announcing the Artificial Analysis Endpoint Accuracy Index, measuring how much of an open weights model's accuracy each serverless API endpoint preserves
SemiAnalysis/Kimbo Chen, Shubham Choudhari, Bryan Shan and Dylan Patel
Kimi K3, The Manos, The Mythos, The Legendos
Kimi K3’s architecture: compressed memory, attention across depth, latent expert routing, and serving performance
Subscribers onlyChips & hardwareInference & pricingModels & releasesRead at newsletter.semianalysis.com
July 2026
Artificial Analysis
DeepSeek V4 Flash 0731 scores 50 on the Artificial Analysis Intelligence Index, 10 points above previous DeepSeek V4 Flash
DeepSeek releases DeepSeek V4 Flash 0731
Artificial Analysis
Inkling Small lands within a point of Inkling on the Artificial Analysis Intelligence Index with less than a third of the parameters
Thinking Machines' new Inkling Small scores 40 on the Artificial Analysis Intelligence Index, within a point of its flagship sibling Inkling with less than a third of the total and active parameters
Evals & benchmarksRead at artificialanalysis.ai
SemiAnalysis/Nicolas Bontigui, Eric (Junqi) Wen, Jeremie Eliahou Ontiveros, Nigel Chiang, Reyk Knuhtsen and Dylan Patel
The Wild Wild West Of LEGO Datacenters
Everyone Says They're Modular, Do The Vendor Claims Hold Up? Zuck's Tents, AWS's Houdini, 60GW+ Modular Capacity Tracked, Full Vendor Landscape Mapping, Vertiv's 2x Content Uplift Per MW
Artificial Analysis
Agnes AI releases Agnes 2.5 Pro Alpha
Agnes 2.5 Pro Alpha
Models & releasesRead at artificialanalysis.ai
METR
How independent researchers could investigate AI propensities after misalignment incidents
AI agents sometimes take sophisticated actions in violation of human intent. We outline the questions that thorough external investigations of these behaviors should answer, the access this might require, and how the resulting findings should be shared.
SemiAnalysis/Bryan Shan, Daniel Nishball, Myron Xie, Wega Chu, Ivan Chiam, Cam Quilici, Cheang Kang Wen, Kimbo Chen, Gerald Wong, Zane Fong, Jordan Nanos and Dylan Patel
Can AMD break the CUDA Moat? AMD Advancing AI 2026
Up to 105% Equity Rebate Discounts for OpenAI, Agentic Kernel Generation, Improvement in Software Quality, Unstable Internal Development Clusters, Helios MI455X Production Ramp Hell
Artificial Analysis
Claude Opus 5: the new leader in agentic knowledge work
Claude Opus 5 is the new leader on our agentic knowledge work benchmark, AA-Briefcase, outperforming Claude Fable 5 by nearly 150 Elo while reducing Cost per Task by 20%
Artificial Analysis
Opus 5: Fable 5 level intelligence at a lower cost per task
Claude Opus 5 is narrowly the most intelligent model on the Artificial Analysis Intelligence Index, offering comparable intelligence to Fable 5 at 26% lower Cost per Task
METR/Tom Cunningham
Metrics of Agent Ability
Tom Cunningham surveys metrics for comparing AI agent capability as performance changes with expenditure and relative to human performance.
Agents & automationRead at metr.org
SemiAnalysis/Alec Ibarra, Bryan Shan, Daniel Nishball, Zane Fong, Cam Quilici, Kimbo Chen, Jordan Nanos and Dylan Patel
Vera Rubin NVL72 vs GB200 NVL72? Inference TCO & Architecture Analysis
Rubin LUT Based Tensor Core, Feynman, Rack Scale, Perf Per MegaWatt, Perf Per Dollar, Software Improvements, Public Rubin Software, PyTorch, vLLM, OpenAI Triton
Artificial Analysis
How Thinking Machines Lab’s Inkling performs on agentic knowledge work
Thinking Machines Lab’s Inkling scores an Elo of 836 on on our agentic knowledge work benchmark AA-Briefcase
METR/Parker Whitfill and Tom Cunningham
The Economics of Recursive Self-Improvement
Parker Whitfill and Tom Cunningham highlight context and takeaways from a new paper modeling how AI may accelerate AI R&D, and whether feedback effects could be strong enough for self-sustaining acceleration.
SemiAnalysis/Wayne Ma, Myron Xie, Julien Martin-Prin, Daniel Nishball and Dylan Patel
Meta’s Infrastructure Team Needs A Culture Reset
Meta Infrastructure has become bloated, with middle managers expending resources on over-engineered technology solutions that lose sight of broader organizational needs.
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
Kimi K3: second only to Fable 5 on AA-Briefcase
Kimi K3 is second only to Fable 5 on AA-Briefcase, our agentic knowledge work benchmark, but costs more than Opus 4.8 to run while averaging nearly an hour per task
Artificial Analysis
Gemini 3.6 Flash and Gemini 3.5 Flash-Lite: Halving Time per Task
Google has released Gemini 3.6 Flash and Gemini 3.5 Flash-Lite. Both halve time per task relative to their predecessors and increase token efficiency, Gemini 3.5 Flash-Lite improves by 11 Intelligence Index points while Gemini 3.6 Flash does not improve in intelligence over 3.5 Flash
METR/Tom Cunningham, Manish Shetty, Vincent Cheng and Nate Rush
Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT
We propose a measure of an AI agent’s optimization ability with an "expenditure horizon." We give an empirical illustration from the NanoGPT speedrun.
Artificial Analysis
Four frontier launches in eight days: six labs now field a model above 50 on the Artificial Analysis Intelligence Index
Yeah. Grok 4.5, GPT-5.6, Muse Spark 1.1, and Kimi K3 all launched within eight days.
Artificial Analysis
Kimi K3 achieves #3 in the Artificial Analysis Intelligence Index, comparable to Opus 4.8 and GPT-5.5
Benchmarks and Analysis of Kimi K3
Artificial Analysis
Thinking Machines has released Inkling, the new leading U.S. open weights model
Thinking Machines has released Inkling, the new leading U.S. open weights model, debuting at 41 on the Artificial Analysis Intelligence Index
Artificial Analysis
How GPT-5.6 Sol, Terra, Luna compare on intelligence vs cost
GPT-5.6 Sol and Luna are ahead of Terra at every point on the Intelligence vs Cost per Task chart. GPT-5.6 Luna stands out as a particularly cost efficient model
Artificial Analysis
Muse Spark 1.1: Meta gains 8 Intelligence Index points in three months
Meta's Muse Spark 1.1 scores 51 on the Artificial Analysis Intelligence Index and is cost and token efficient compared to its peers
SemiAnalysis/Max Kan, Julien Martin-Prin, Jeremie Eliahou Ontiveros and Dylan Patel
The Future of Meta Superintelligence: A 1 Year Progress Update
A top tier RL environment startup spawns out of thin air, the most aggressive compute ramp we've ever seen, 2000km+ scale-across, and some advice for Google DeepMind
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
GPT-5.6 benchmarks across Intelligence, Speed and Cost
GPT-5.6 Sol comes close second to Claude Fable 5 in the Artificial Analysis Intelligence Index at one third of the cost, and leads the Artificial Analysis Coding Agent Index in OpenAI’s Codex harness
Artificial Analysis
Announcing EnterpriseOps-Gym-AA
Our independent leaderboard for ServiceNow’s EnterpriseOps-Gym
Evals & benchmarksRead at artificialanalysis.ai
Artificial Analysis
Grok 4.5 brings SpaceXAI to the intelligence frontier
SpaceXAI’s Grok 4.5 scores 54 to place fourth on the Artificial Analysis Intelligence Index following only Fable 5, GPT-5.5, and Opus 4.8. It scores on par with GPT-5.5 in Codex on the Artificial Analysis Coding Agent Index in the Grok Build harness, at much lower cost
METR/Thomas Kwa
Because 8 ≈ e², Anthropic's researcher uplift is plausibly >2x
Anthropic reported that contributors merge 8x as much code per day as before AI. Thomas Kwa argues that under standard economic modeling assumptions, this implies the uplift of individual Anthropic researchers from coding agents alone is above 2x.
SemiAnalysis/Joey Brookhart, Crystal Huang and Dylan Patel
Anthropic 3Q26 Profit Over $1B: The Anthropic IPO Financials Sneak Peak
Anthropic’s Opportunity is Theirs to Lose
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
Announcing Harvey LAB-AA: evaluating AI agents on real-world legal work
Our implementation of Harvey's Legal Agent Benchmark tests language models on 120 private legal tasks spanning 24 practice areas, graded against rubrics of binary criteria.
SemiAnalysis/Daniel Nishball, Cheang Kang Wen, Zane Fong, Reyk Knuhtsen, Sebastian Orejas, Jordan Nanos, Terence Ong, Dylan Patel and Hemanth Ajithkumar
Nvidia GPU Debt Backstop Unleashes the AI Project Trinity: Capital, Offtake and Datacenters
Over 7T AI debt by 2029, There can be no Neoclouds without the Trinity. Nvidia's Backstop Economics Explained.
Subscribers onlyChips & hardwareDatacenters & energyEconomics & laborRead at newsletter.semianalysis.com
Artificial Analysis
Announcing AutomationBench-AA
Our independent leaderboard for Zapier’s AutomationBench
SemiAnalysis/Jeremie Eliahou Ontiveros, Max Kan, Joey Brookhart, Crystal Huang and Dylan Patel
Meta Compute: Everyone Wants To Be A Neocloud
Zuck Takes Plan B? SpaceX 2.0, Bedrock 2.0, MSL Isn't Giving Up, Scaling RecSys by 10x...
Subscribers onlyRead at newsletter.semianalysis.com
SemiAnalysis/Afzal Ahmad, TC, Gerald Wong and Dylan Patel
EMIB-T Roadmap, Custom HBM, HBM4 Packaging Challenges, Microfluidic Cooling, Photonic Interconnects, and More
ECTC 2026 Roundup, Intel, TSMC, SK Hynix, Samsung, Micron, Marvell, Lightmatter, Microsoft
June 2026
SemiAnalysis/Crystal Huang, Joey Brookhart and Dylan Patel
TokenBudgeting: Our Conversations with Enterprises on Token Spend
Was Widespread TokenMaxxing Ever Really Here?
Artificial Analysis
Claude Sonnet 5: strong agentic performance at a higher cost per task
Claude Sonnet 5 achieves 53 on the Artificial Analysis Intelligence Index, but without promotional pricing will cost more per task than Opus 4.8
METR
Summary of METR's predeployment evaluation of GPT-5.6 Sol
A summary of METR's independent, predeployment evaluation of GPT-5.6 Sol
SemiAnalysis/Jeremie Eliahou Ontiveros, Sebastian Orejas, Ellie Holbrook and Dylan Patel
US Grid Constraints: Towards 40GW+ of Behind-The-Meter Datacenter by 2028?
Why the Grid Can't Keep Up, and Why that Drives Behind-The-Meter 50%+ of DCs/Year By 2028
Artificial Analysis
Measuring time per task in AA-Briefcase
Agentic knowledge work can take frontier models over 20 minutes per task, as measured in AA-Briefcase, our new benchmark
SemiAnalysis/Ray Wang, Myron Xie, Dylan Patel, Junsung Kim, Sravan Kundojjala and Louis Lu
China’s CXMT Is Set to Challenge DRAM Incumbents
CXMT IPO, SK Hynix, Micron, Samsung Competition, Process Node Deficit, China HBM, Wafer Adds, Memory LTAs
Artificial Analysis
Announcing the Artificial Analysis Speech to Speech Index
Announcing the Artificial Analysis Speech to Speech Index, our new synthesis metric for native Speech to Speech model quality, comprising of Big Bench Audio, Full Duplex Bench, and 𝜏-Voice
SemiAnalysis/Reyk Knuhtsen, Maya Barkin, Jeremie Eliahou Ontiveros, Ellie Holbrook, Nicolas Bontigui and Dylan Patel
Stop Saying Half of 2026 US Datacenter Capacity Is Canceled
Don't Believe The Vibecoded Estimates, Go Through Every Individual Filing Instead
Artificial Analysis
Announcing AA-Briefcase: a frontier knowledge work evaluation
AA-Briefcase is a new benchmark for testing models on realistic knowledge work tasks in complex projects built by industry experts. Models are evaluated on multi-week knowledge work projects, each with many linked tasks and thousands of input source files, combining rubric and pairwise grading to evaluate verifiable…
SemiAnalysis/Kimbo Chen, Cheang Kang Wen and Dylan Patel
RL Systems Mind the Gap: Matching Trainer and Generator Throughput
RL Training Infrastructure, GRPO, PipelineRL, Async RL, Policy Staleness, RL Sandbox Infra, CPU Requirements, TCO Analysis, Thinking Machines Tinker
Artificial Analysis
GLM-5.2 is the new leading open weights model on the Artificial Analysis Intelligence Index
Benchmarks and Analysis of GLM-5.2
Artificial Analysis
Artificial Analysis Intelligence Index v4.1: a shift toward agentic workloads
Announcing Artificial Analysis Intelligence Index v4.1: a shift toward agentic workloads, featuring upgraded benchmarks and new per-task metrics
SemiAnalysis/STEEL Team, Afzal Ahmad, Andrew Wagner, Gerald Wong, Wayne Ma, Dylan Patel, Daniel Sanchez, Adith Shankar, Allison Elliott and Sarah Lawrence
Is SMIC N+3’s Metal Pitch Smaller than Intel 18A’s?
SMIC N+3 Node Deep Dive vs TSMC N6, TechInsights Private Equity Sale, SemiAnalysis Teardown Engineering & Evaluation Lab, HiSilicon Kirin 9030, Process Technology, Patterning, Cell Architecture
Artificial Analysis
First results from AA-AgentPerf: the hardware benchmark for the agent era
AA-AgentPerf measures how many concurrent agents an AI system can serve on real coding-agent trajectories while meeting production service-level targets, with Agents per Megawatt as its lead metric. The first results cover NVIDIA and AMD systems, from single accelerators to full racks.
SemiAnalysis/Doug, Sravan Kundojjala and Dylan Patel
Intel Should Raise Capital
Intel's woes are behind them. The heavy spending is ahead of them.
Artificial Analysis
Benchmarking guardrail models for safety, refusal, and latency
An Artificial Analysis benchmark report for guardrail safety models.
SemiAnalysis/Bryan Shan, Cam Quilici, Kimbo Chen, Alec Ibarra, Dylan Patel, Daniel Nishball and Cheang Kang Wen
DeepSeekV4 1.6T Day 0 to Day 43 Performance Over Time - Huawei, GB300 NVL72, MI355X, B200
Day 0 Inference Performance, InferenceX, 100x performance improvement in 26 Days, Cost per Million Tokens, Huawei 950DT Inference Trace Analysis
Artificial Analysis
Claude Fable 5 Launches at #1 on the Artificial Analysis Intelligence Index
Anthropic is nearly 5 points ahead of any other lab’s best model
Artificial Analysis
Claude Fable 5: the first public Mythos-class model
Anthropic has released Claude Fable 5, the first publicly available Mythos-class model that ranks #1 in our agentic real-world knowledge work benchmark GDPval-AA
Artificial Analysis
North Mini Code: Cohere's small coding-focused MoE model
Cohere just released North Mini Code, a small 30B parameter (3B active) open weights coding model that scores 27.6 on the Artificial Analysis Intelligence Index
SemiAnalysis/Reyk Knuhtsen, Niko Ciminelli, Jacob Rintamaki, Robert Ghilduta, Joe Ryu, Jeremie Eliahou Ontiveros and Dylan Patel
Unitree's Impossible Trajectory Is Still Overlooked
The Fastest Iteration Cycle In Next-Gen Robotics Should See Unprecedented Acceleration
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
MiniMax-M3: Leading open weights model, once the weights are released
MiniMax-M3 scores 55 on the Artificial Analysis Intelligence Index. Once the weights are released, it will be the leading open weights model
Artificial Analysis
NVIDIA Nemotron 3 Ultra released: fast, intelligent, and open
NVIDIA released Nemotron 3 Ultra today - the most intelligent open weights model from a US lab
SemiAnalysis/Daniel Nishball, Pranav Myana, Ellie Holbrook, Harley Blackard, Zane Fong, Cheang Kang Wen, Sravan Kundojjala, Tanj Bennett, Myron Xie, Terence Ong and Hemanth Ajithkumar
To Boldly Go: The Case for Space Datacenters
Space DC Total Cost of Ownership Explained. Unpacking constraints from Terrestrial DCs and Chip Production.
Subscribers onlyChips & hardwareDatacenters & energyInference & pricingRead at newsletter.semianalysis.com
Artificial Analysis
Fun-Realtime-TTS: New Text to Speech model topping Artificial Analysis leaderboard
Alibaba's Fun-Realtime-TTS takes the #1 spot on the Artificial Analysis Speech Arena Leaderboard, surpassing Google's Gemini 3.1 Flash TTS and Inworld's Realtime TTS-2 Research Preview
Artificial Analysis
MAI-Transcribe-1.5: New Speech to Text model leading the accuracy-speed Pareto frontier
Microsoft has released MAI-Transcribe-1.5: an exceptionally fast speech transcription model at a speed factor of ~276x, while still achieving 2.4% on AA-WER (#3), leading the accuracy-speed Pareto frontier
Artificial Analysis
AA-WER Streaming: New Speech to Text Streaming Benchmark
Announcing AA-WER Streaming, our new benchmark measuring streaming Speech to Text models on accuracy and latency for voice agent use cases. Pareto optimal models on this new benchmark include those from Cartesia, ElevenLabs, and Deepgram
May 2026
Artificial Analysis
Nemotron 3 Ultra announced: high-speed, leading US open weights intelligence
NVIDIA just announced the release of Nemotron 3 Ultra in Jensen Huang's Computex keynote: at 550B parameters (55B active), this is the largest Nemotron 3 model to date, and it is the most intelligent US open weights model
SemiAnalysis/Malcolm Spittler and Dylan Patel
AI Dark Output: The Visible Cost of Invisible Output
Why AI's increasing output is going to be one of the hardest economic measurement problems in history. AI "Dark Output" could end up being the majority of economic activity, but a challenge to measure
Artificial Analysis
Claude Opus 4.8 - The new #1 AI model
Anthropic retakes #1 on GDPval-AA and advances in terminal use and scientific reasoning
Models & releasesRead at artificialanalysis.ai
SemiAnalysis/Justin Lebar
Finding Miscompiles for Fun, Not Profit
Or: You don’t need access to Claude Mythos to spend $10,000 in an afternoon.
SemiAnalysis/Jeremie Eliahou Ontiveros, Joey Brookhart, Crystal Huang and Dylan Patel
Anthropic Growth and Bedrock Mix Drive AWS Margins Higher While Peers Lag
Amazon’s Bedrock Mix and Anthropic Deal Terms Combine to Show Greater Operating Leverage
Subscribers onlyRead at newsletter.semianalysis.com
Artificial Analysis
ITBench-AA: Agentic benchmark to test models on real-world SRE tasks
Artificial Analysis and IBM are launching ITBench-AA, the first in a new series of benchmarks evaluating models on agentic enterprise IT tasks, starting with Site Reliability Engineering tasks where frontier models score below 50%
Artificial Analysis
MiniCPM5-1B: The leading 1B open weights model
OpenBMB has released MiniCPM5-1B (Non-reasoning), the leading 1B open weights model scoring 17.9 on the Artificial Analysis Intelligence Index
SemiAnalysis/Nicolas Bontigui, Jeremie Eliahou Ontiveros, Konrad Wang, Aran Industries, Derek Yin and Dylan Patel
Inside the 800VDC Revolution – Part 1
Four-Phase 800VDC Transition, Power Rack Economics, SST, Equipment Content/MW Build, Supplier Implications
SemiAnalysis/Sravan Kundojjala, Dylan Patel and Gerald Wong
EDA Market Primer - Market Dynamics, Cadence, Synopsys, Siemens, China EDA Rise
EDA Market size, Share, Business Models, Drivers, Changing Customer Base, Competitive Dynamics, IP, Hardware, CoT, Lock-In Economics, Disruptive Forces
Subscribers onlyModels & releasesPolicy & geopoliticsEconomics & laborRead at newsletter.semianalysis.com
Artificial Analysis
Cursor’s Composer 2.5: third on the Coding Agent Index and ~10-60x lower cost than rivals
This release puts Composer among the leading coding agent models, something that wasn’t clear for past releases
Artificial Analysis
Cohere launches open weights model Command A+, more than a year since the Command A release
Benchmarks and analysis of Command A+
METR
Frontier Risk Report (February to March 2026)
A pilot assessment of rogue deployment risk at frontier AI companies. Starting in February 2026, METR conducted a pilot exercise to assess misalignment risks from AI agents used inside frontier AI developers, with participation from Anthropic, Google, Meta, and OpenAI.
Artificial Analysis
Gemini 3.5 Flash: The new leader in intelligence versus speed
Benchmarks and analysis of Gemini 3.5 Flash