New models, version upgrades, open weights and capability descriptions.
LatestUnsloth releases GGUF version of DeepSeek-V4.1-Flash-1017
Why read thisUnsloth has released a GGUF quantized version of DeepSeek-V4.1-Flash-1017, along with a 64GB memory requirement, making it easier for local deployers to evaluate.
Why read thisThe original text provides the technical report, the open-source benchmark, and a comparison video with Genie 4, so readers can judge the reproducibility of this world model.
Baseten's Base Labs has released joey-1, a 2B decision model that directly outputs a single-token label and calibrated confidence, with a P50 end-to-end latency of 20ms, about 2.5 times faster than elephant-1 (49ms), at a price of $0.01 per million tokens, one-tenth that of mammoth-1. joey-1 is post-trained from Qwen3.5-2B and shares the same data recipe and API with elephant-1 and mammoth-1, so switching requires changing only one line. The model is now available on Baseten Model APIs, with zero data retention by default, and its weights are open-sourced on Hugging Face; Model APIs supports a 16k context window, while dedicated deployments can unlock a 262k context window and support fine-tuned weights.
Why read thisThe original text provides joey-1's latency, pricing, and the division of labor among the three model tiers, so readers can judge whether decision-making tasks should switch to a smaller model.
NVIDIA has released the open-weight model Nemotron 3.5 Transcribe on Hugging Face and ModelScope, integrating transcription, speaker diarization, translation, and audio question answering into a single checkpoint. It supports transcription in 40 languages, translation between 31 languages and English, and speaker labeling in 13 languages, can process recordings up to 1 hour long, and can label up to 8 speakers. The model ranks first among open-weight models on the English leaderboard of the Hugging Face Open ASR Leaderboard with an average WER of 5.33%, and achieves 2.8% AA-WER on Artificial Analysis with approximately 4B activated parameters. The simultaneously released Granary v2 covers 48 languages and 463,062 hours of speech data, and the accompanying NeMo Speech fine-tuning recipe reduced WER on held-out speakers from 22.44% to 12.17% in a medical case.
Why read thisThe original text provides unified weights for multilingual transcription, speaker diarization, and translation, along with public leaderboards and fine-tuning recipes, making it easy to assess deployment approaches.
Mistral has released the open weights for Mistral Large 4 (Le Chonk), a 1-trillion-parameter natively multimodal model that continues reinforcement learning on top of the preview version, with improved coding and defensive cybersecurity capabilities. It scores 19.7% on FrontierCode 2.0 and 48% on the Artificial Analysis Coding Agent Index, roughly on par with GLM-5.3, but uses an average of 47% fewer tokens per task; on Terminal-Bench 4.0 it improves by 13 percentage points to 41%. The model supports three reasoning modes, uses a modified MIT-based Mistral Ecosystem Open-Weight license, and is priced via API at $0.70 per million input tokens, $0.07 for cached input, and $2.10 for output.
Why read thisThe original text provides open weights, three reasoning modes, and comparisons across multiple benchmarks, allowing readers to judge the trade-offs for self-hosted coding and security scenarios.
Microsoft AI has released the image editing model MAI Image 2.7, launched in preview form, ranking 3rd on Arena Edit and 2nd on the Artificial Analysis image editing leaderboard, making Microsoft AI the 2nd-ranked lab. The company states that the model outperforms 2.6 on all 10 use cases and 7 categories of editing actions evaluated by Artificial Analysis, leading the next-best model by 24 Elo in scenes and styles. It is currently available for trial on Arena, Artificial Analysis, and MAI Playground, with text-to-image functionality coming soon.
Why read thisThe official rankings of this image editing model on Arena Edit and Artificial Analysis leaderboards are given, along with the improvement margin of use cases relative to 2.6.
Liquid AI has released the LFM3 series of open-weight models, all branched from LFM3-1.7B-Base, covering text, vision, audio, Omni, and the d1 decision model. LFM3-1.7B-Base adopts a three-hybrid-layer architecture, is pretrained on 22T tokens, and in the mid-training stage over the final 3.5T tokens incorporates more agent trajectories and tool-calling data, with the context window extended to 256K tokens. The company says LFM3-1.7B is about 25% faster than LFM2-1.2B at an 8K context, and scores 38.42 versus 3.02 on WebArena-Verified; LFM3-VL-2B's average score across 62 vision benchmarks rises from 57.6 to 71.4. All models are available for download on Hugging Face, along with a single-GPU fine-tuning notebook.
Why read thisThe original article presents the architectural changes shared by the six models on a common backbone and on-device benchmark data, which can be used to assess the trade-offs of the small-model specialization approach on long-context and agentic tasks.
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.
Why read thisThe original text explains that Genie 4 adds native audio and natural language control to real-time interactive simulation, allowing readers to understand the new capability boundaries of world models.
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.
Why read thisThe original text provides Carbon's benchmark results on cybersecurity and long-horizon coding tasks and its phased rollout path, which can be used to assess the pace at which frontier models will be deployed in attack and defense scenarios.
Alphabet's Q3 2026 revenue grew 25% year-over-year, with annualized revenue reaching $500 billion, and Google Cloud revenue up 103%. CEO Sundar Pichai said first-party model APIs process nearly 32 billion tokens per minute, 40% more than last quarter; the Gemini app has reached 1 billion users, making it the fastest-growing product in Google's history. A new model in the Gemini 4 series, Carbon, was also released that day; it ranks first on the Artificial Analysis Intelligence Index with a score of 59, will first be made available to select trusted cybersecurity defenders, and is being tested by the U.S. government.
Why read thisThe earnings call provided specific figures on the Gemini 4 series, model API call volume, and cloud business growth, which can be used to gauge the current pace of Google's AI commercialization.
OpenAI announced that its model achieved three mathematical and theoretical physics breakthroughs in one shot: proving the local smoothing conjecture in all dimensions, from which it follows that the Kakeya conjecture and the Stein restriction conjecture hold in all dimensions; proving that the Exponential Time Hypothesis (ETH) is false; and constructing existence proofs for quantum chromodynamics and Yang-Mills theory. The relevant proofs have been formalized using Lean and uploaded to a public GitHub repository. OpenAI has also invited external teams, including Russell Impagliazzo and Paturi, the original proposers of ETH, and Nobel laureate David Gross, to participate in the review, and is not seeking academic authorship.
Why read thisThe original article outlines the proof paths and external review arrangements for OpenAI's three mathematical results, which can be used to observe the verification mechanisms after AI produces scientific conclusions.
Why read thisThe official release presents the SR-SWA mechanism and results on two long-document benchmarks, offering insight into the context and latency trade-offs of long-document OCR.
Why read thisOpenRouter launches GPT-6.1 Terra, offering key comparisons: the same score as Astra, one-tenth the price, and three times the speed.
Why read thisThe official announcement sets out the new model's positioning in visual reasoning and computer operation, and explains the scope of access for paid users.
Yanxin Matrix, under RockAI Technology, together with universities including HKUST and CMU, has released and open-sourced EdgeRock-4B. At a 256K context, its KV Cache is fixed at 72MiB, a 99.1% reduction compared with the source model Qwen3.5-4B. A single 80GB GPU can run 512 concurrent long-inference streams, with an output throughput of 12,124 tokens/s, 6.42 times that of the source model; the average pass@1 on AIME 2024–2026 is 82.9%. The model is derived from Qwen3.5-4B through staged conversion, requiring no pretraining from scratch; the weights and inference code have been open-sourced and are available for commercial use, supporting text and image inference.
Why read thisThe original text provides measured comparisons of fixed memory and throughput, based on which readers can assess changes in memory cost for long-inference scenarios.
Dream Labs has released the world-action model Dream-1, which can control unseen dual-arm parallel-jaw gripper robots to perform unseen tasks via a public API, without fine-tuning or demonstration data from that robot. The company also announced it is emerging from stealth with a $90 million seed round. Dream-1 costs $3 per hour to run, and Dream-1-flash costs $2 per hour; each call returns 1.6 seconds of actions. If zero-shot performance is insufficient, it can be fine-tuned via FaaS with 5 minutes of robot data.
Why read thisDream Labs has disclosed Dream-1's zero-shot performance on unseen robots and its hourly pricing, which can be used to gauge the barrier to deployment for robot foundation models.
Block's internal AI research lab released the full-duplex speech-to-speech model Dessa on October 23, stating that it was trained entirely from scratch on human conversational speech, can listen and speak simultaneously, and natively perceives emotion, sarcasm, and laughter. The model's inference speed is faster than real time, and it can run locally on a MacBook Pro M4 Max or higher configuration; the team completed training with a single NVIDIA GB200 NVL72. Block has published demo recordings and a public waitlist, and has opened early access to a small number of AI safety researchers.
Why read thisDessa was trained from scratch using only speech data, demonstrating how full-duplex conversation handles emotion and interruption timing, and also providing local running requirements.
Why read thisKimi K3.1 Pro's Code Arena ranking and cost data, directly comparable with Claude Opus-5 and the previous-generation K3.
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.