New models and versions, open weights, and changes in capability and pricing, with their evaluations.
Google released its next-generation flagship model, Gemini 4 Argon. It scored 77.9% on DeepSWE v1.1, ahead of Claude Opus 5.5 at 74.2% and GPT-6 Astra at 74.1%, and ranked first on Text Arena with 1,525 points.
Why it matters: Google's new flagship tops several leaderboards and publishes cost and maximum output-token figures, helping readers assess its position against GPT and Opus.
Also, 6.1 Sol is our fastest-growing model ever. It was a bit slow under heavy load before. It should be a lot better now!
AI2 released Olmo-core 3, a major upgrade to its large language model development framework. It redesigns the open MoE training system to scale MoE training to trillions of parameters while maintaining compute efficiency.
Why it matters: The source provides throughput and memory comparisons for the MoE training stack, helping readers assess changes in the cost of large-scale MoE training.
SynthID Bio is our new family of watermarking methods designed for AI-generated biological designs. A world first: we can now embed an imperceptible signature directly into protein sequences without affecting their biological function. 🧵
Google DeepMind released Gemini 4 Argon for coding, enterprise knowledge work, and cyber defense. It is initially available to government users and trusted cybersecurity professionals in the Fairwind program, before expanding to developers, businesses, and consumers.
Why it matters: This summary combines Gemini 4 Argon's benchmark results, pricing, and differences in third-party evaluations, enabling comparison with Astra and Opus 5.5.
Google parent Alphabet released Gemini 4 Argon, saying it can handle coding, research, and writing and was specifically trained for defensive cybersecurity. It can autonomously find, verify, and patch critical software vulnerabilities.
Google released frontier model Gemini 4 Argon, matching OpenAI and Anthropic on several benchmarks without a clear lead. It is initially available to 'trusted cyber defenders' in the Fairwind program and Google's internal teams before expanding to developers, businesses, and consumers. Promotional API pricing is $2 per million input tokens and $10 per million output tokens.
Why it matters: This summary compares third-party and human-preference leaderboards, scores, token use, and pricing for Gemini 4 Argon against OpenAI and Anthropic.
Demis Hassabis announced the new frontier AI model Gemini 4 Argon, saying it delivers major improvements in several key capabilities. Starting today, it is being responsibly opened through the Fairwind Program to government users and trusted cybersecurity defenders, with availability to expand later.
Why it matters: The release announcement gives Gemini 4 Argon's capability positioning and phased access plan, showing the rollout pace for a frontier model.
Google released frontier model Gemini 4 Argon, describing frontier performance in software engineering, enterprise knowledge work such as legal and financial tasks, and cyber defense. It is initially available only to a group of trusted cyber defenders. Google says it is participating in the U.S. government's pre-release frontier-model access process and will gradually expand access, while strengthening safeguards against misuse, prompt injection, and misalignment before a wider rollout.
Why it matters: Google released Gemini 4 Argon initially only to trusted cyber defenders, reflecting a trend of tying frontier model releases to security review.
Google DeepMind released frontier model Gemini 4 Argon, initially making it available through the Fairwind Program to trusted cybersecurity defenders before gradually expanding access. Pricing starts at $2 per million input tokens and $10 per million output tokens.
Why it matters: Gemini 4 Argon raises the maximum output to 1M tokens and reports multiple benchmark results and internal deployment data, offering a view of its limits on long-running workflows.
Gemini 4 Argon!
Google released Gemini 4 Argon, claiming industry-leading performance in coding, knowledge work, and cybersecurity, but for now it is available only to company engineers, with no public access.
Google released frontier model Gemini 4 Argon, initially opening it to cyber defenders today and planning to expand access as soon as possible. Introductory pricing is $2 for input and $10 for output.
Why it matters: The source gives the new model name, initial audience, and pricing, helping readers assess its capabilities and cost.
Google DeepMind released frontier model Gemini 4 Argon for complex workflows including coding, enterprise knowledge work, and cyber defense. It is available starting today to a group of trusted testers through the Fairwind Program.
Google announced Gemini 4 Argon, a new frontier model built to sustain deep reasoning across complex, long-horizon workflows, with frontier performance in software engineering, enterprise knowledge work such as legal and finance, and cybersecurity defense. Google is expanding the model's output token limit to 1M tokens. Argon is rolling out to trusted cyber defenders in the Fairwind Program, with broader availability planned as soon as possible.
We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view. pplx-embed-v2-context-9b-preview sets a new state of the art on ConTEB and @turbopuffer's new, privately held context-bench. https://www.perplexity.ai/hub/blog/contextual-embedding-beyond-the-gold-passage
Google DeepMind launched SynthID Bio, bringing watermarking to synthetic biology. It embeds imperceptible signatures directly into biological code so that generated content can be verified both in digital models and in synthesized physical proteins, while preserving their biological function.
Why it matters: SynthID watermarking has expanded from images to synthetic biology, showing how AI-generated proteins can be marked and verified.
OpenAI released its next-generation flagship model GPT-6 Astra at 1 a.m. Beijing time on September 4. Leiphone compiled first-hand tests from users who received access early.
Why it matters: Early user tests show GPT-6 Astra can check its own work on long tasks but still needs human oversight, revealing its capability limits.
StepFun released Step 5 Preview, scoring 44 on the Artificial Analysis Intelligence Index and ranking among the top two open models globally.
Leiphone compared HiDream-O1-Video-1.0, Seedance 2.5, and MiniMax H3 on three original image-to-video tasks, evaluating intent planning, temporal causality, and audio-video alignment.
The author tested Qwen3.8-Flash and GLM-5.3-Flash, released on the same day, in Claude Code on the same task: building a personal workspace from scratch with tasks, calendar, notes, a dashboard, and global search.
Qwen3.8-27B is now available through Nebius Token Factory. The company says it targets coding, deep research, and agent workflows, with an emphasis on planning and completing multi-step tasks. It is a 27B dense model accessible to developers through a Nebius Token Factory endpoint.
Anthropic released Claude Sonnet 5.5. API prices are unchanged and generation is faster, while its scores rose substantially on agent benchmarks such as Terminal-Bench and CursorBench.
On September 28, Alibaba's Qwen-Audio-3.1-TTS speech model ranked first globally on the Controlled Voice Arena leaderboard by Artificial Analysis with an Elo rating of 1,177.
AWS announced that Claude Sonnet 5.5 is available on Amazon Bedrock and Claude Platform on AWS. It is positioned as a more efficient Sonnet model for coding and knowledge work, with lower per-task costs and faster speed for most tasks.
Why it matters: The source describes Sonnet 5.5 improvements for coding and document tasks, its division of work with Opus 5.5, and access through Bedrock.
ElevenLabs released voice model Eleven v4, which better follows emotion, pauses, and sound-effect tags in scripts and keeps voices consistent in long productions. The new architecture also powers Turbo, a variant for real-time voice agents. In company tests, it begins speaking in about 150 milliseconds, compared with 262 milliseconds for Cartesia Sonic 3.6 and 814 milliseconds for OpenAI GPT-4o mini TTS.
OpenAI launched GPT-6.1 Sol at DevDay, saying it approaches GPT-6 Astra’s intelligence on agentic coding, computer use, and professional work, at one-fifth the standard input and output token price.
Why it matters: OpenAI launched GPT-6.1 Sol at DevDay at one-fifth the price, while also disclosing that GPT-6.1 Astra had been held back over safety concerns.
In a September 29 post, Simon Willison discussed GPT-6.1 Sol’s near-Astra intelligence at about one-fifth the price. The original post did not disclose specific model parameters, benchmark scores, or pricing figures.
Sam Altman shared the introduction of 6.1 Sol, saying it offers intelligence close to Astra at one-fifth the price, with a 95% discount on cached reads. The post also said Ultrafast can make Astra eight times faster and will support 6.1 Sol as well.
OpenAI launched GPT-6.1 Sol, saying it delivers intelligence close to Astra at one-fifth the price and is currently the most cost-efficient model at comparable performance.
Why it matters: OpenAI’s pricing and performance positioning for GPT-6.1 Sol helps readers assess its value against current models.
OpenAI has released GPT-6.1 Sol, saying it approaches the flagship GPT-6.1 Astra—which was delayed over safety concerns—in agentic coding, computer use, and office tasks, at about one-fifth the cost.
Why it matters: With flagship Astra delayed over safety concerns, readers can compare Sol and Astra with Opus 5.5 on price and benchmark gaps.
NVIDIA has released Kumo Tabular, an open-source tabular foundation model that predicts labels for new rows in a single forward pass from labeled table rows, supporting classification and regression without training, tuning, or feature engineering. The model comes in three sizes from 28M to 215M parameters, was pretrained entirely on synthetic tables, is commercially usable under the OpenMDW-1.1 license, and ranks first on four benchmarks: TabArena, BeyondArena, TALENT, and ScoringBench.
Why it matters: The source reports model sizes, training-data provenance, and rankings on four benchmarks, helping readers assess the accuracy and efficiency tradeoffs of tabular foundation models.
Microsoft Research has released Quine, an AI research system designed for biological complexity. It combines a biological world model with an interactive harness connecting models, scientific tools, literature, and experimentalists. In work with the Broad Institute on pancreatic ductal adenocarcinoma, it predicted and prioritized thousands of compounds; the top-ranked compound produced the largest classical-to-ground-state transition in wet-lab experiments. The process from narrowing the search space to selecting a candidate for validation took one weekend.
Why it matters: Microsoft Research describes Quine's multimodal biological world model and experimental loop, including wet-lab validation in pancreatic cancer compound screening.
OpenAI has released GPT-6.1 Sol, positioning it as near-Astra intelligence for coding, computer use, and professional work. Standard API input and output tokens cost one-fifth as much as Astra.
Why it matters: OpenAI describes GPT-6.1 Sol as close to Astra in capability at one-fifth the API price, giving readers a basis for weighing cost against performance.
IQuest Research at Zhizhi Innovation Institute has released IQuest-Q1, a decoder-only Transformer with a sparse mixture-of-experts architecture, about 320B total parameters and about 15B active parameters. It is available on GitHub and Hugging Face.
Latent Space's AINews roundup of developments from September 24–25 says community feedback was positive after Opus 5.5 launched that week, especially for generating explainer videos with code.
Why it matters: The weekly roundup offers a quick look at model and agent-infrastructure developments, including community feedback on Opus 5.5's code-generated videos and how vendors position their new models.
Anthropic has released Claude Sonnet 5.5. The company says it runs more than 30% faster and costs up to 30% less on most tasks. It retains Sonnet 5's pricing while outperforming its predecessor across benchmarks.
Why it matters: The author tests Sonnet 5.5's coding performance and availability on the free tier, and compares it directly with Opus 5.5.
Mike Krieger shares @claudeai’s announcement of Claude Sonnet 5.5, the second model in the Claude 5.5 family. It runs more than 30% faster than Sonnet 5 and costs up to 30% less for most work.
Why it matters: The comparison with Sonnet 5 gives readers a concrete reference for the release’s reported speed improvement and cost reduction for most work.
Anthropic has announced the release of Claude Sonnet 5.5, the second model in the Claude 5.5 family. The company says it runs more than 30% faster than Sonnet 5 and costs up to 30% less for most work.
Why it matters: Anthropic's official launch of the second Claude 5.5 model gives comparable figures for changes in speed and cost.
Over the past six months… - Muse Spark - Muse Image - Muse Video - Muse Spark 1.1 - Meta Model API - Muse Glimmer - Muse Spark 1.2 - Muse Code - Muse Spark 1.3 - Muse …and we're just getting started. 🚀
