New models and versions, open weights, and changes in capability and pricing, with their evaluations.
OpenAI announced that GPT-6 is rolling out globally in ChatGPT alongside Intelligent UI. The company says the new version responds faster and provides visual content and interactive experiences users can explore and use directly.
Why it matters: OpenAI announced the global rollout of GPT-6 in ChatGPT alongside Intelligent UI, showing how the model and interaction format are changing together.
AWS announced that Claude Haiku 5.5 is available on Amazon Bedrock and Claude Platform on AWS.
Why it matters: AWS describes Haiku 5.5's positioning, cost changes, and invocation methods, helping readers assess its role in a tiered multi-agent setup.
Mike Krieger highlights Claude Haiku 5.5 as built for speed and efficiency, costing 75% less than Haiku 4.5. He describes Opus handling heavy thinking and Haiku handling high-volume work. The quoted @claudeai announcement calls it its cheapest, fastest, and most capable small model, with average running costs around 75% lower than Claude Haiku 4.5. https://twitter.com/claudeai/status/2107894039626277339
Why it matters: The comparison with Haiku 4.5 gives a concrete cost reference for high-volume work, alongside the author's suggested division of labor between Opus and Haiku.
Anthropic released its latest model, Claude Haiku 5.5. According to a post by @claudeai cited in the source, it is the company's cheapest, fastest, and most capable small model so far, with average operating costs about 75% lower than Claude Haiku 4.5.
Why it matters: Anthropic officially released Haiku 5.5, highlighting an approximately 75% reduction in operating costs compared with the previous generation.
Anthropic introduced Claude Haiku 5.5, describing it as the cheapest, fastest, and most capable small model it has released. The company says running it costs around 75% less on average than Claude Haiku 4.5. No pricing figures, benchmarks, or availability details were given in the post.
Why it matters: Anthropic's own announcement states the cost and speed positioning of Claude Haiku 5.5 relative to Haiku 4.5, giving readers the vendor's headline claims for the small model tier.
Liquid AI released two open-weight models in its d1 decision-model family: d1-3B and the experimental d1-omni-600M.
Why it matters: Liquid AI open-sourced two decision models and reported on-device latency and results across seven benchmarks, providing evidence on the use of small models for structured decisions.
Perplexity is releasing pplx-embed-v2-late, described as two late-interaction embedding models that retrieve text, images, and pages through a shared embedding space for cross-modal querying. The company says both models reach frontier performance and are publicly available on Hugging Face, with a blog post at perplexity.ai/hub/blog/multimodal-embeddings-beyond-a-single-vector.
Mistral released its 1-trillion-parameter Mistral Large 4, codenamed Le Chonk. Anyone can use and customize it for free during the preview; the final version is due by the end of the month. The model is for general tasks, with additional optimization for coding, cyber defense, and fields such as manufacturing, finance, and electrical engineering. Mistral says Le Chonk is the strongest open-weight model outside China so far and comes close to some closed models. The company emphasized that it was trained from scratch rather than distilled.
Why it matters: Mistral is using a 1-trillion-parameter open-weight model to target specialized areas such as coding and cyber defense, offering a view of how the gap between open and closed models is changing.
Mistral AI released Mistral Large 4. The company says it can run general-purpose agents that gather information and produce finished deliverables in complex workflows.
EmbeddingGemma 2 uses the Apache 2.0 license. Simon Willison argues that embedding models should not be locked into proprietary hosted services. They often require computing and storing thousands or millions of vectors; if a provider discontinues a model, users may have to pay to recompute them. He would rather pay for a hosted version while retaining the option to run open weights or switch providers if the service shuts down.
Google DeepMind released EmbeddingGemma 2, a 740M-parameter open model based on the Gemma 4 architecture and licensed under Apache 2.0. It maps text, images, audio, and video into a shared embedding space.
Why it matters: EmbeddingGemma 2 maps text, images, audio, and video into one embedding space and reports concrete on-device memory and vector truncation figures, helping assess local multimodal retrieval.
Musubi released PolicyLM-1.7B, a lightweight decision model with open weights that applies content policies written in plain English to messages in 50 milliseconds. Its cost and speed are close to the AI classifiers used for moderation by major social platforms, while it offers the flexibility of a modern LLM and needs no specialized training to follow complex policies. Policy changes do not require retraining. Musubi co-founder and chief AI officer Filip Jankovic said this lets platform managers label content proactively.
Mistral released a preview of Mistral Large 4, a model with 1 trillion total parameters and 49 billion active parameters. It was trained on the company's own cluster of 3,800 NVIDIA Grace Blackwell GPUs and is currently available through its API.
Google released Nano Banana 2.1, an image-generation and editing model that replaces Nano Banana 2, released in February 2026. Prices are about half as much: a 1K image drops from 6.70 cents to 3.36 cents, and a 4K image from 15.10 cents to 7.56 cents.
Google released EmbeddingGemma 2, an open multimodal embedding model that converts text, images, video, audio, and code into numerical vectors. It has 740 million parameters. Google says it outperforms competitors with more than twice as many parameters on multimodal embedding benchmarks and scores 78.68 on the Massive Text Embedding Benchmark (Code), nearly 10 points above the previous model's 68.76.
Simon Willison used the same prompt to ask Mistral Large 4, claude-opus-5.5, gpt-6.1-sol, and gemini-3.8-flash each to generate an SVG of an armadillo jaywalking on Mars in fishnet stockings, comparing frontier models after benchmark saturation. He ran the test with the LLM command-line tool, using each model's default reasoning level.
Google released Nano Banana 2.1 for image generation and editing. It delivers higher-quality images, fixes several issues in the previous model, and costs less. It is available through the API, AI Studio, and the Gemini app.
Mistral AI says Mistral Large 4 is one of the world's strongest AI models for cybersecurity.
Google DeepMind released EmbeddingGemma 2, which it calls the first natively multimodal open embedding model for on-device use. It combines code, images, audio, and video with text in a shared space.
Sundar Pichai announced EmbeddingGemma 2, described as Google's first open, natively multimodal embedding model. It handles text, code, image, video, and audio tasks in a modular 740M parameter form factor, is positioned for offline, privacy-first RAG when paired with Gemma 4, and is said to outperform some specialist models more than twice its size. Weights are available now on Hugging Face.
French AI lab Mistral AI released Mistral Large 4 (ML4) on Tuesday, a 1-trillion-parameter multimodal model nicknamed Le Chonk. It is currently accessible only through a guarded public endpoint, with open weights planned within three weeks after safety testing is complete.
Mistral released its largest model to date, Mistral Large 4, with 1 trillion parameters and 49 billion active parameters. An API preview is available in Mistral Studio; weights are expected at the end of October.
Mistral released a public preview of Mistral Large 4, a natively multimodal model with 1 trillion parameters and 49 billion active parameters. Its weights will be opened at the end of the month.
Why it matters: Mistral outlined the open-weight release schedule for its 1-trillion-parameter multimodal model and its investment in European compute, allowing comparison with mainstream open models.
Mistral AI released Mistral Large 4, codenamed Le Chonk, with 1T total parameters, native multimodality, and 49B active parameters. The company says it is the best open-weight model on aggregated benchmarks in Europe and the U.S. It reaches state of the art on workloads in cyber defense, manufacturing, and finance, and exceeds closed frontier models in visual grounding. Built end-to-end in Europe, it can be deployed from Europe through Mistral Cloud. The API is available today, and open weights are due at the end of October.
Why it matters: Mistral specified a 1T-parameter, 49B-active open-weight model and a local European deployment path, offering a view of the open-weight field's scale and availability.
AI company Reflection released its first free, open model, Beam, targeting coding, logical reasoning, and agent tasks. It uses a mixture-of-experts architecture, activating 23 billion of its 501 billion total parameters per token.
Falcon-Emirati-7B is a specialized model for Emirati dialects built on Falcon-H1-Arabic. It aims to understand and generate Emirati Arabic like a native speaker. It uses the Falcon-H1 hybrid architecture, running Mamba alongside Transformer attention. The base model family comes in 3B, 7B, and 34B sizes, with a context window of up to 256K tokens.
Reflection released Beam, a text-only mixture-of-experts model for coding, agents, and scientific work. It has 501B total parameters and 23B active parameters, was trained from scratch, and will release all weights under Apache 2.0 later this month.
Zhipu AI (Z.ai)'s GLM 5.3 is now available on Amazon Bedrock. The 753B-parameter mixture-of-experts model targets coding and long-horizon agent tasks and is offered through a fully managed API with cross-region inference, prompt caching, and service tiers. It is currently available to eligible enterprise customers.
Why it matters: The source describes how to access GLM 5.3 on Bedrock and use prompt caching, helping readers assess the cost impact of managed inference for coding and long-horizon agent workflows.
Reflection AI officially released its first frontier open-weight model, Beam, saying it performs comparably to leading Chinese open models on advanced reasoning benchmarks with 3–4 times less inference compute. Beam is a 501-billion-parameter, 23-billion-active-parameter text model, pretrained on 23.8 trillion tokens with a 1-million-token context window. The company says it matches Z.ai's GLM-5.2 and exceeds current leading Western open models.
Reflection AI introduced Beam, described as a highly efficient agentic open model with 501B total parameters and 23B active parameters. The company says it advances the Western open frontier on coding and agentic tasks and was trained end-to-end from scratch, with full weights to be released this month. Ollama said more US models are coming to its platform.
Why it matters: Reflection AI's Beam is presented as an efficient agentic open model with 501B total and 23B active parameters, giving readers a concrete comparison point for open-weight coding and agentic models.
Reka AI released a research preview of Rho-1, a 19B-parameter general-purpose model that can process and generate text, images, video, and robot-control actions in a single neural network. Unlike most systems that route tasks to specialized models, Rho-1 places all modalities as tokens in one shared context window without calling tools or external models. It can generate continuous video in real time, respond instantly to new instructions without restarting, and use the same weights that predict camera images to drive robot actions.
Aleph Alpha released Kolibri, a bilingual German-English model with 78B total parameters using a mixture-of-experts architecture that activates about 3B parameters per token. The company says Kolibri sits on the Pareto frontier for quality and operating cost in both languages. It scores 71% on German benchmarks and decodes faster than comparable models including GPT OSS A5B, Qwen 3.6 A3B, and Gemma 4 A4B.
NASA and IBM Research, together with several academic institutions, released the NASA-IBM Lunar Foundation Model, which they describe as one of the first open-source foundation models for lunar science.
Cloudflare released Clef and Clef-flash, two decision models for AI agents. They return brief classifications with probabilities rather than long text, and the company says agent decisions no longer need human involvement.
Ai2 open-sourced AstaBrief 8B, a model that turns research questions and retrieved literature excerpts into reports with citations. It was trained from Qwen3-8B using SFT and DPO and is now available as Asta's Fast mode.
Why it matters: The source describes AstaBrief's training-data construction and citation-filtering details, which can be adapted to deploy other long-form generation models.
Microsoft AI released MAI-Transcribe-2-Streaming, a real-time transcription model supporting 60 languages. Its first partial results arrive in just over 100 milliseconds. Microsoft says it ranks first for accuracy on Artificial Analysis; audio costs $0.54 per hour through the end of the year.
Black Forest Labs released Flux 3 Image, an image model in the Flux 3 family. The company says it supports multi-step editing without changing other parts of an image and covers text-to-image, image-to-image, text rendering, and photorealistic generation.
Why it matters: Flux 3 Image combines multi-step editing with 4K output and a reference-image limit, allowing comparison with Ideogram 4.5 on the pace of competition in image-editing models.
Tavus launched Griffin, which the company calls the first Human Interaction Model (HIM). In live video calls, it can receive and generate speech, facial expressions, tone of voice, gestures, and pauses at the same time.
Ideogram released Ideogram 4.5, which the company says changes only the area specified by the user while leaving the rest intact, avoiding artifacts after repeated edits. It offers four quality tiers priced from $0.008 to $0.22 per image, all in native 2K resolution. The model is available on Ideogram's platform and API, with partners including Picsart, Runway, Pika, and Leonardo AI. The company says an open-weight version is coming.
Microsoft AI is launching MAI-Transcribe-2-Streaming, which Mustafa Suleyman calls the world's most accurate real-time transcription model. He claims it is 55% faster and 60% cheaper than ElevenLabs and invites developers to build agents on Microsoft's platform. In the quoted post, Artificial Analysis reports #1 rankings for final and first partial transcript accuracy on AA-WER Streaming, with final transcript WER of 2.5% at 0.13s after speech ends and streaming pricing of $0.54 per hour of audio.
Why it matters: The quoted benchmark comparison pairs transcription accuracy with latency and pricing, giving readers concrete tradeoffs to consider when choosing a streaming speech model for agents.
