Skip to content
K = 0.731
2026.10.17 · Sat

#编码

#编码

Models

New models and versions, open weights, and changes in capability and pricing, with their evaluations.

10/16Fri
  1. Hugging Face Blog83

    Reflection releases Beam, a 501B open model for coding and agentic work

    Reflection released Beam on Hugging Face, a 501B MoE model with 23B active parameters per token, a 1M-token context window, and an Apache 2.0 license. The text-only model targets coding and agentic work, with BF16, FP8, and NVFP4 checkpoints and day-0 support in transformers, vLLM, and SGLang. Hugging Face provides deployment and post-training recipes alongside its own task evaluations and demos.

    Why it matters: The deployment guidance distinguishes active computation from total weight memory, helping teams assess hardware requirements for running and adapting a sparse model with a long context window.

  2. Hugging Face Blog87

    Meta releases Muse Spark Open, an open-weight multimodal mixture-of-experts model

    Meta’s Muse Spark Open is an open-weight mixture-of-experts model with approximately 816 billion parameters, around 42 billion active per token, and a context window of up to one million tokens. Trained for agentic coding, it accepts text, images, audio, and video and generates text. Hugging Face provides inference and deployment recipes, reports its own quantization tests, and demonstrates attention-layer LoRA fine-tuning with TRL while keeping the experts frozen.

    Why it matters: The deployment recipes clarify how sparse expert activation reduces per-token computation while retaining substantial memory requirements, helping developers assess hardware needs across checkpoint formats and serving frameworks.

10/14Wed
  1. 宝玉 · X85

    Google releases Gemini 4 Argon with paid access and introductory API pricing

    Baoyu reports that Google released Gemini 4 Argon on October 13 for complex, multistep tasks, citing Sundar Pichai’s announcement of global paid availability. Access includes the paid Gemini API and platform, Ultra subscriptions in Google Antigravity, and Pro or Ultra subscriptions in the Gemini app and Search. Introductory API pricing is $2 per million input tokens and $10 per million output tokens, including thinking tokens, through November 30; prices double December 1. Input, thinking, and output share a 1,048,576-token limit.

    Why it matters: The shared budget for input, thinking, and output makes context allocation consequential for long tasks, while repeated input accounting leaves a specific billing detail unresolved.

  2. Google DeepMind94

    Google DeepMind launches Gemini 4 Argon for complex reasoning and agentic tasks

    Google DeepMind has launched Gemini 4 Argon globally through its API and selected Google products, with access depending on subscription or service tier. Introductory pricing is $2 per million input tokens and $10 per million output tokens, with cached input tokens at 95% off and a 1M unified input-output token limit through the Gemini API. Google reports improvements across coding, professional workflows, scientific research, and robotics, alongside stronger agent safeguards and prompt injection resistance.

    Why it matters: Google’s reported gains in code submissions and reductions in rollbacks connect Argon’s engineering capabilities to delivery speed and code stability in internal workflows.

  3. Logan Kilpatrick · X86

    Google rolls out Gemini 4 Argon across Gemini API, Gemini App, and Antigravity

    Logan Kilpatrick announces that Gemini 4 Argon is rolling out globally today in the Gemini API, to Pro + Ultra users in Gemini App, and to Ultra subscribers in Antigravity. He says over 100,000 people at Google tested the model over the last month, with cyber defenders in the fairwind program and early access customers receiving it over the last 2 weeks. He highlights writing, knowledge, persistence for coding, and 1M token output when the model needs to work more.

    Why it matters: The rollout details clarify how access differs across Gemini API, Gemini App, and Antigravity, helping readers understand where they can use the new model.

10/13Tue
  1. Latent Space84

    Meta launches Muse Wonder and Muse Wonder Full frontier models

    Meta Superintelligence Labs released Muse Wonder and Muse Wonder Full, its second-generation model family for personal and professional agentic work. Latent Space reports that Wonder is available through the Meta Model API and Muse Code, while Full is limited to Muse Code subscribers. Artificial Analysis gives Wonder Full and Wonder Intelligence Index v4.3 scores of 56 and 52, respectively, versus Opus 5.5’s 58; Wonder supports a 1M context window and text, image, video, and audio inputs, with text-only output.

    Why it matters: The benchmark breakdown compares Wonder variants with competing models on agentic performance and token usage, showing how their relative standing changes across tasks rather than a single ranking.

  2. Alexandr Wang · X82

    Meta releases Muse Wonder and Muse Wonder Full with higher benchmark scores

    Meta has released Muse Wonder and Muse Wonder Full, according to Artificial Analysis in the post quoted by Alexandr Wang. At max reasoning effort, they score 52 and 56 on the Artificial Analysis Intelligence Index v4.3, compared with Muse Spark 1.3’s 48, following a release gap of 6 weeks. Muse Wonder is available through Meta’s first-party API and Muse Code, while Muse Wonder Full is limited to Muse Code users; both have 1M-token context windows.

    Why it matters: Artificial Analysis’s comparison highlights a tradeoff between Muse Wonder’s improved benchmark performance and its higher cost per task relative to Muse Spark 1.3 despite lower output token usage.

  3. TechCrunch · AI83

    Meta releases frontier models Muse Wonder and Muse Wonder Full

    Meta’s Superintelligence Lab released Muse Wonder and Muse Wonder Full on Monday, larger models described as more capable than Muse Spark. Meta reports benchmark scores comparable to Fable 5.1 and GPT-6 Astra and emphasizes information retention through “active memory.” Wonder costs $4.00 per million input tokens and $20.00 per million output tokens through the Meta Model API; Wonder Full is available only to Muse Code subscribers. Both models will power the Muse agent.

    Why it matters: The split between API access for Wonder and subscriber access for Wonder Full gives developers a concrete basis for comparing integration options within Meta’s new model lineup.

10/8Thu
  1. Latent Space82

    Anthropic releases Claude Haiku 5.5 at GPT-6 Luna pricing

    Anthropic released Claude Haiku 5.5, the first Haiku-series update in about a year. Its pricing matches OpenAI's GPT-6 Luna. On the same day, Anthropic also cut prices for Sonnet 5.5 and its subscription plans.

    Why it matters: The report brings together Haiku 5.5 pricing, independent evaluations, and token-usage data to help assess cost tradeoffs for small models in agent workflows.

10/6Tue
  1. Mistral AI80

    Mistral launches public preview of Mistral Large 4, a 1-trillion-parameter multimodal model

    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.

  2. AWS Machine Learning Blog62

    GLM 5.3 launches on Amazon Bedrock

    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.

  3. Ollama · X65

    Reflection AI announces Beam, a 501B-parameter agentic open model

    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.

10/1Thu
  1. Google DeepMind88

    Google DeepMind releases frontier model Gemini 4 Argon

    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.

9/30Wed
  1. 雷锋网78

    First hands-on tests of GPT-6 Astra: build Manhattan in a week, but long tasks still fail

    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.

9/29Tue
  1. The Decoder80

    OpenAI releases GPT-6.1 Sol at one-fifth the cost of Astra

    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.

  2. OpenAI News78

    OpenAI releases GPT-6.1 Sol

    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.

  3. Latent Space78

    AINews: Opus 5.5 excels at generating explainer videos

    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.

  4. AWS Machine Learning Blog62

    Claude Sonnet 5.5 launches on Amazon Bedrock and Claude Platform on AWS

    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.

9/23Wed
  1. Latent Space88

    Anthropic releases Claude Opus 5.5 as OpenAI launches GPT-6 Sol and Luna the same day

    Anthropic has released Claude Opus 5.5, the first model in its new Claude 5.5 family. The company says it reaches Claude Fable 5.1 levels on most tasks, costs 40% less to run than Opus 5, and becomes the default model in Claude Code and the Claude app.

    Why it matters: Two labs launched new models and cut prices sharply on the same day, giving readers a way to compare their approaches to capability and per-task cost.

9/3Thu
  1. Demis Hassabis · X78

    Google releases Gemini 3.8 Flash and 3.8 Flash Cyber

    Google has released Gemini 3.8 Flash, another upgrade less than a month after the previous one, and introduced Gemini 3.8 Flash Cyber for cybersecurity. Demis Hassabis says Flash Cyber advances the frontier of cyber defense; more details about the models and security work are on Google's official blog. Logan Kilpatrick says the model improves agentic and coding capabilities, marking the Flash line's third update in six weeks.

    Why it matters: Google has updated the Flash family again in under a month and introduced a dedicated Cyber version, showing its model-iteration pace and security focus.