NAll news

All news

Wednesday44 items

  1. Google BlogFirst-partyModel releases
    AI score52

    Google Releases Gemini 4 Argon, Showcasing Ability to Handle 8 Categories of Complex Tasks

    Google has released its latest model, Gemini 4 Argon, saying it is aimed at complex real-world tasks and can perform reasoning, coding, and coordinate tools. The company provided 8 examples, including writing 5,000 lines of mathematical code from scratch with zero source material to render a 3D sunlit library, and, in Antigravity, processing more than 80,000 raw files from a NASA Mars mission and generating executive briefing slides.

    Collage of various uses of Gemini 4 Argon

    #Product updates#Multimodal#Reasoning

  2. Google DeepMindFirst-partyModel releases
    AI score77

    Google DeepMind releases world model Genie 4, supporting native audio and real-time interaction

    Google DeepMind has released the world model Genie 4, which can generate interactive worlds in real time and, for the first time, natively generate ambient sound and character dialogue. The model builds on capabilities from the Gemini series; Gemini 3.8 Flash breaks down goals in a loop and checks frames to maintain causal coherence across multi-step events, and Gemini 4 Argon has also been used for joint simulation tasks with Gemini Robotics. The company says it remains an early experimental model: over long runs, object permanence and complex physical interactions still drift, all generated content carries a SynthID watermark, and it is currently available in preview to trusted testers first.

    Same story · 28 reports

    #Model releases#Multimodal#Video

  3. Google BlogFirst-partyModel releases
    AI score78

    Google DeepMind releases world model Genie 4, supporting real-time interactive simulation and native audio

    Google DeepMind releases the world model Genie 4, focused on real-time interactive simulation and introducing native audio for the first time in the series. The model is built on capabilities from the Gemini family: Gemini Omni is responsible for making simulations match the appearance and behavior of the real world, while other Gemini models provide the complex reasoning needed to maintain simulation coherence. Users can control scenes in real time using natural language, such as moving the camera, changing the weather, or introducing new characters, and each world also generates ambient sound, music, and interactive dialogue on the fly.

    Same story · 28 reports

    #Model releases#Multimodal#Google

  4. MarkTechPostModel releases
    AI score66

    Liquid AI Releases the LFM3 Series of 6 Open-Source 2B-Class Multimodal Models

    Liquid AI has released the LFM3 series, in which 6 open-weight models share a single backbone and cover text, vision, audio, omni-modal, and decision-making tasks, all of which can be fine-tuned on a single GPU. LFM3-1.7B has 1.73B parameters and a 256K context, requires about 1 GB of memory at 4-bit, and scores an average of 49.37 across 29 benchmarks in official evaluations, higher than Qwen3.5-9B's 47.24. The weights are now available on Hugging Face, support formats such as GGUF, MLX, and ONNX, and the threshold for free commercial use is annual revenue of $10 million.

    Same story · 4 reports

    #Model releases#Open-source ecosystem#Multimodal

  5. Google DeepMindFirst-partyModel releases
    AI score87

    Google DeepMind Releases Gemini 4 Carbon, Focused on Cybersecurity and Long-Horizon Task Capabilities

    Google DeepMind has released Gemini 4 Carbon, calling it its strongest model, ranking first with 59 points on the Artificial Analysis Intelligence Index v5.0, and achieving 88.2% on SEC-bench Pro, 54.8% on ExploitGym, and 71.0% on CWE-bench. The model is aimed at long-horizon engineering tasks; an internal team once had it run continuously for several days to investigate and fix hidden defects in a large codebase, and it rose to 4th place on WebDev Arena with 1,768 points. Starting this week, Carbon is first being made available to select cybersecurity defenders, then gradually expanding through the Fairwind Program, and only afterward to developers, enterprises, and consumers; after the introductory period, pricing is $4 per 1 million input tokens and $20 per 1 million output tokens.

    Faceted iridescent crystal surrounded by floating white number 4s.

    Same story · 34 reports

    #Model releases#Agents#English text is being prepared.

  6. Google BlogFirst-partyModel releases
    AI score82

    Google DeepMind Releases Gemini 4 Carbon Frontier Model

    Google DeepMind has released Gemini 4 Carbon, calling it its strongest model, ranking first with 59 points on the Artificial Analysis Intelligence Index v5.0 and leading on SEC-bench Pro (88.2%), ExploitGym (54.8%), and CWE-bench (71.0%). The model targets long-horizon engineering tasks; an internal team once had it run continuously for several days to investigate and fix hidden defects in a large codebase, and it rose 89 elo to 4th place (1768) on WebDev Arena. Starting this week, Carbon will first be opened to some cybersecurity defenders, then gradually expanded through the Fairwind Program, and only afterward made available to developers, enterprises, and consumers; after the introductory period, pricing will be $4 per 1M input tokens and $20 per 1M output tokens.

    Faceted iridescent crystal surrounded by floating white number 4s.

    Same story · 34 reports

    #Model releases#Safety / alignment#Agents

Tuesday46 items

  1. t3nModel releases
    AI score85

    Moonshot AI Releases Open-Weight Model Kimi K3.1 Pro

    Moonshot AI has released the open-weight model Kimi K3.1 Pro, a post-trained version of Kimi K3 that scores 65.7% on Terminal-Bench 4.0, higher than GPT-6 Astra and slightly lower than Claude Opus 5.5. Artificial Analysis calls it the strongest open-weight model currently available, with an intelligence index of 55 points, but notes that generation speed is slow and output is verbose; in evaluation it generated 150 million tokens, compared with a median of 81 million. Pricing is $2 per million input tokens and $10 per million output tokens. Moonshot AI plans to release the full weights by November 4 and says it has restricted the model's offensive cybersecurity capabilities.

    Same story · 34 reports

    #Model releases#Open-source ecosystem#Safety / alignment

  2. MarkTechPostModel releases
    AI score60

    Mirendil Releases Its First Model Fern 0.1, Supporting a 10M Token Context

    Mirendil has released its first model, Fern 0.1, a coding model trained from scratch with a context window of up to 10M tokens, equivalent to roughly 23 MB of code, or about 500,000 lines. Mirendil-Base was pretrained by a single researcher in 3.5 months, with a total compute cost of less than $10 million; AI agents wrote more than 99% of the code, and the training data came only from public sources. RULER single-needle retrieval scored 100% at every length from 32K to 10M tokens, but Terminal Bench 4.0 scored only 6.1%, and SWE-Bench Verified scored 73.4%. Currently only hosted early access is available, and the weights have not been open-sourced.

    Same story · 2 reports

    #Model releases#English text is being prepared.#Agents

  3. Analytics India MagazineModel releases
    AI score62

    Maya Research Releases Open-Weight Speech Model Maya 2 Native Supporting 12 Languages

    Maya Research has released Maya 2 Native, an open-weight text-to-speech model that supports 12 languages, including 11 Indian languages and English, under the Apache 2.0 license. The model supports real-time audio streaming output, generating 24 kHz mono audio in 80-millisecond frames, and comes with 20 built-in synthetic voices; voice cloning is possible using approximately 3 to 10 seconds of reference audio. Running it locally requires an NVIDIA GPU, and the documentation recommends at least 12 GB of VRAM per session; the model card also notes limitations with mixed-language text, URLs, numbers, and some cross-lingual voice combinations, and that audio watermarking is not included.

    #Model releases#Speech#Open-source ecosystem

  4. ComputerworldModel releases
    AI score75

    OpenAI releases the GPT-6.1 Terra reasoning model, priced 50% lower than GPT-5.6 Terra

    OpenAI has released the reasoning model GPT-6.1 Terra, replacing GPT-5.6 Terra, the predecessor of GPT-6.1 Terra, at a 50% lower price, positioned between GPT-6 Luna and GPT-6.1 Sol and GPT-6 Astra. On Agents' Last Exam, it scores 60.0% at the highest effort level at a cost of $1.01 per task, versus 50.7% and $3.58 for its predecessor; on OSWorld 2.1 it scores 74.2%, still below Claude Sonnet 5.5's 83.9%. API pricing is $1 per million input tokens, $5 for output, and $0.05 for cached input, and it is now available to Plus, Pro, Enterprise, and Edu users of ChatGPT Work and Codex.

    Same story · 32 reports

    #Model releases#Reasoning#OpenAI

  5. Rohan PaulModel releases
    AI score52

    Hi3D Open-Sources the Hi3D v1.0 Image-to-3D Model

    Hi3D has open-sourced the former Sparc3D under the name Hi3D v1.0. This is an image-to-3D model, and its code and weights have been released on GitHub and Hugging Face and can be used commercially. The model takes an image and a coarse mesh generated by another model, refines it in three passes at three resolution levels of 512, 1024, and 1536, uses sparse voxels and a self-pruning decoder, and has expanded its training data from 500K to about 2M 3D assets. Among 8 compared models, its multi-view fidelity of 0.75 and detail richness of 0.841 are both the highest, its 3D print slicing success rate is 90%, and in 562 side-by-side ratings by 13 professional 3D experts, 83.5% chose it; the model outputs only shape without color or texture, requires an NVIDIA GPU, and is free for commercial use up to 200,000 monthly active users or $1 million in monthly revenue.

    Quoted postHi3D · @Hi3D_AI

    https://t.co/FlZfsBfs0T

    Same story · 2 reports

    #English text is being prepared.#Image generation#Hugging Face

  7. Shao MengModel releases
    AI score61

    Baidu Open-Sources Unlimited OCR 2: 3B/A0.5B MoE Architecture Supports Ultra-Long Document Parsing

    Baidu has open-sourced Unlimited OCR 2, an OCR and layout parsing model designed for documents ranging from dozens to hundreds of pages, adopting a 3B/A0.5B MoE architecture. Compared with v1's R-SWA, v2 upgrades the decoder attention to SR-SWA and adds Unlimited-Horizon Decoding, which compresses the attention interaction between text and vision into a fixed-capacity working reference set (the README gives 512 tokens as an example), and then combines this with page-by-page chunked decoding, enabling ultra-long documents to be parsed within a limited context. The authors claim that at the hundred-page scale, step-by-step decoding speed is approximately 7.0× that of Full Attention and approximately 5.6× that of R-SWA, with latency and GPU memory barely increasing as documents grow longer.

    Quoted postBaidu Inc. · @Baidu_Inc

    Unlimited OCR 2 is here, and it's open source. 📖 ✨ Longer documents can mean a lot more attention computation. Our new SR-SWA mechanism keeps that in check by condensing text-visual attention interactions into a fixed-size set of working references. Here's what it delivers: - Document parsing at virtually any length, within a finite context window - Nearly constant per-step decoding latency and memory footprint as documents grow longer - State-of-the-art performance on MPDocBench and MarathonDocBench, covering documents from dozens to hundreds of pages Check out the model here ↓

    Same story · 5 reports

    #Model releases#Open-source ecosystem#Multimodal

  8. t3nModel releases
    AI score71

    OpenAI Releases GPT-6.1 Terra Mid-Tier Model

    OpenAI has released GPT-6.1 Terra, a mid-tier model positioned between performance and cost, available to Plus, Pro, Enterprise, and Edu subscribers and already integrated into ChatGPT Work and Codex. According to OpenAI, the model partially surpasses GPT-6 Astra on communication and computer-operation tasks at a fraction of the latter's cost; in tests designed to deliberately induce errors, its error rate was 16.7%, compared with 53.6% for the previous-generation GPT-5.6 Terra. API pricing is $1 per million input tokens and $5 per million output tokens.

    Same story · 32 reports

    #Model releases#OpenAI#Reasoning

  9. heise onlineModel releases
    AI score80

    OpenAI releases GPT-6.1 Terra, focused on low-cost computer task automation

    OpenAI has released GPT-6.1 Terra, positioned as a model for fast, low-cost automation of computer workflows, sitting between Sol and Luna in the GPT-6 family, with Astra remaining the flagship. According to OpenAI's measurements, on the medium reasoning tier of the Agents' Last Exam benchmark, Terra is 15 percentage points higher than the previous generation GPT-5.6 Terra, with a per-task cost of about one-third of the previous generation, reaching the level of GPT-6 Astra, while Astra's API token price is ten times higher. Independent testing by Artificial Analysis shows that Terra solves about 56% of tasks on Terminal-Bench 4.0 at the highest reasoning tier (35% for the previous generation), lower than Claude Sonnet 5.5's approximately 64%; the API price is $1 per million input tokens and $5 per million output tokens, half that of GPT-6.1 Sol and Claude Sonnet 5.5. The model will gradually be made available to Plus, Pro, Business, Enterprise, and Edu users in ChatGPT Work and Codex, and can also be used via the API; it is not offered in regular ChatGPT chat.

    Same story · 32 reports

    #Model releases#Agents#English text is being prepared.

  10. The ElecModel releases
    AI score82

    Moonshot AI Releases Kimi K3.1 Pro, AA Intelligence Index Score of 55 Surpasses GPT-6 Astra

    Moonshot AI has released Kimi K3.1 Pro, an upgraded version of Kimi K3, featuring a MoE architecture with 2.8 trillion total parameters and 104 billion activated parameters, activating 16 of 896 experts at a time. According to Artificial Analysis benchmarks, the model ranks fourth on the Intelligence Index with a score of 55, higher than GPT-6 Astra's 53, while the previous generation Kimi K3 scored 44. It ranks first on Terminal Bench 4.0 with 67%, compared with 13% for the previous generation and 64% for Claude Sonnet 5.5; pricing is $2 per million input tokens and $10 per million output tokens, but its output speed of 73 tokens/s is below the average of 81 tokens/s.

    Same story · 6 reports

    #Model releases#Reasoning#Agents

  11. LeiphoneModel releases
    AI score72

    Baidu Open-Sources ERNIE Unlimited OCR 2, Supporting Document Parsing of Any Page Count

    Baidu has released and open-sourced ERNIE Unlimited OCR 2, claiming it is the first to achieve document parsing of any page count; in testing, it parsed all 7 books of the Harry Potter series in a single pass, totaling more than 3,600 pages. The model achieved overall scores of 84.6 and 77.1 on MPDocBench and MarathonDocBench, respectively. It adopts the Sparse Reference Sliding Window Attention (SR-SWA) mechanism, and in hundred-page document scenarios, its single-step decoding speed is 7 times and 5.6 times faster than Full Attention and the previous-generation R-SWA, respectively. The code and model weights have been open-sourced on GitHub and Hugging Face, and the model is also available on the Baidu AI Cloud platform and the Baidu AI Open Platform.

    Same story · 5 reports

    #Model releases#Open-source ecosystem#Multimodal

  12. gihyo.jpModel releases
    AI score90

    OpenAI releases GPT-6.1 Terra, with API pricing at half that of GPT-5.6 Terra

    On October 26, 2026, OpenAI released and made available GPT-6.1 Terra, strengthening visual understanding, computer operation, and writing capabilities. The standard API price is half that of GPT-5.6 Terra, namely $1 per million tokens for input, $0.05 for cached input, and $5 for output. The model has a context window of 1.05 million tokens, a maximum input of 922,000 tokens, and a maximum output of 128,000 tokens. Reasoning effort can be selected from none to max, with medium as the default. On Agents' Last Exam, the medium setting scores 15 points higher than GPT-5.6 Terra while costing about one-third as much; under the max setting, the average speed for completing tasks is more than 3 times that of GPT-6 Astra and GPT-6.1 Sol. The model is already available to Plus, Pro, Enterprise, and Edu users of ChatGPT Work and Codex, and the API model ID is gpt-6.1-terra.

    Same story · 32 reports

    #Model releases#Agents#Reasoning

  13. Shanghai ObserverModel releases
    AI score86

    OpenAI Releases GPT-6.1 Terra, Standard API Price Cut 50% Versus Previous Generation

    OpenAI has released a new model, GPT-6.1 Terra, with standard API input pricing reduced to $1 per million tokens and output at $5, a 50% decrease compared with GPT-5.6 Terra. The model focuses on visual understanding and computer operation, scoring 50.3% at the highest reasoning effort on the Terminal-Bench 4.0 coding benchmark, up from 21.5% for the previous generation. OpenAI says the most complex coding tasks are still better suited to GPT-6.1 Sol or GPT-6 Astra.

    Same story · 32 reports

    #Model releases#OpenAI#Agents

  14. YicaiModel releases
    AI score83

    On the same day Anthropic filed for its IPO, OpenAI released GPT-6.1 Terra and...

    On October 26 local time, on the same day Anthropic submitted its IPO registration filing to the SEC, OpenAI released its new model GPT-6.1 Terra, with standard API prices cut by 50% compared with the previous generation. The model focuses on visual understanding and computer operation, with standard API pricing at $1 per million input tokens, $5 for output, and $0.05 for cached input; on the Terminal-Bench 4.0 coding benchmark, it scored 50.3% at the highest reasoning effort, compared with 21.5% for the previous generation. OpenAI said Terra has approached flagship-model levels on some tasks, but the most complex coding tasks are still better suited to GPT-6.1 Sol or GPT-6 Astra.

    Same story · 32 reports

    #Model releases#OpenAI#Agents