Transformers releases6565OpenBMB's MiniCPM-OCR model released v5.21.0 version, using SigLIP visual encoder and Qwen3.5 language model, specifically for reading scanned pages and screenshots.
Microsoft Research5858Flint-interactive makes AI-generated charts highly interactive, allowing users to explore data through clicks, filtering, zooming, highlighting, and cross-chart interactions. It simplifies the construction of complex interactions through advanced semantic events and update operations, and leverages the Flint compiler to handle interaction logic and visual dependencies, enhancing reliability and development efficiency. Flint-interactive supports creating interactive data stories, dashboards, and connected experiences, enabling AI agents to generate charts with interactive features to help users analyze data more efficiently.
The DecoderFeaturedSame story · 2 outlets7474AI startup Recursive has released its first language model, Flint. During the development process, the AI system named Eureka Machine participated in proposing ideas, building, testing, and planning subsequent experiments.
Why it matters: Recursive publicly announces its first model developed with the assistance of an automated research system, allowing readers to observe how far AI-assisted R&D can go.
MarkTechPostRecursive releases the open-source 40B fuzzy model Flint, 2.5B active parameters, performs best among MoE models below 50B, supports inference and OpenAI compatible tool calls, 4-bit weights about 23GB, SWE-bench Verified 74.2%.
Ollama releases7070Recursive Flint v0.40.8 released, bringing code intelligence agents to Ollama. The 40B parameter MoE model supports 262K token context and achieves 54.9% on Terminal-Bench 2.1 and 74.2% on SWE-bench Verified. Flint can be used for code understanding, debugging, feature implementation, and testing, and supports MacOS.
TechCrunch · AISame story · 2 outlets7171Recursive releases its first model, Flint, a free coding model that can run on laptops, with almost the entire development process driven by AI. Recursive's automated research system, Eureka Machine, participated in almost every development step of Flint, enabling the team to complete the model construction in less than four months.
Ollama releases3535Ollama v0.40.8-rc0 has been released, adding support for the Flint model. It retains the source tool JSON and decoding API values to maintain numeric types and pattern order in prompts, while keeping the existing API serialization unchanged.
OpenHands BlogSame story · 2 outlets6363Recursive Flint is released, a hybrid expert model with 4 billion parameters, designed for efficient local inference, activating only about 250 million parameters per token, achieving 125+ tokens/second on M4 Pro MacBook Pro, and reaching 150+ tokens/second on NVIDIA RTX Spark. OpenHands, as its first partner, supports the model's operation on local hardware, providing developers with an open-source coding model option.
vLLM BlogFeaturedSame story · 2 outlets6060vLLM announced support on the day Flint was released, with both NVFP4 and BF16 checkpoints capable of running on a single GPU, decoding approximately 500 tokens/s per request on a single GB200 with a context of 100,000 tokens.
Why it matters: The original text provided Flint's day-0 support on vLLM, single-card throughput data, and the step-by-step kernel optimization process, allowing readers to judge the engineering trade-offs of small-batch inference based on this.
Ollama releases4040Speech Ollama can use Phonon to generate speech on Apple Silicon, streaming audio via the /v1/audio/speech endpoint; the d3 decision model supports parameter sizes from 2B to 27B for classification, scoring, and other tasks; Flint expanded to Linux and Windows platforms, other improvements include error reporting and model capability transfer.
Edge0 open-sourced the Edge0-145B and Edge0-120B two edge models, and also released a unified inference framework and deployment tools. Officially, the 145B model can run locally on compatible laptops, with peak memory of 29GB, decoding speed of 80+ tokens/s, pre-filling speed of 1100+ tokens/s, supporting context lengths of 128K–256K.
Why it matters: The original text provides actual memory, speed, and multiple benchmark comparisons of the edge 145B model, which can be used to judge the practical boundaries of running cutting-edge models locally.
O'Reilly Radar2222Claude Security reveals increased AI security risks, open-source model competition accelerates, Kimi 3.1 Pro becomes a leader and gets an accelerated version by Moonshot AI; Meta releases Muse series and OpenAI Decisions API.
vLLM releasesvLLM v0.33.0 released, including 927 commits and support from 389 contributors, new 501B MoE Reflection Beam-1.0 and its DFlash2 renderer.
vLLM releasesvLLM v0.33.0 released, adding 927 commits, 389 contributors, introducing 10 new models such as Reflection Beam-1.0 (501B MoE, 23B active) and Yandex AliceAI-Foundation-80B-A3B
Latent SpaceSame story · 3 outlets6565Anthropic has stopped traditional coding interviews for managers, believing Claude has performed better in React and assembly language tests, and points out that the most successful builders do not depend on these test performances. It suggests that junior engineers should prioritize business impact rather than backlog tasks, propose radical ideas, and build prototypes over the weekend, focusing on a good idea and leveraging fresh perspectives.
How the accounts differ:Latent Space states Anthropic stopped coding interviews because Claude performed better in React and assembly language tests, while The Decoder focuses on AI's ability to handle programming tasks better than candidates.
Ollama releasesLiquid AI releases the LFM3 small model family, including the 1.7B parameter lfm3 text model (256K token context), lfm3-vl supporting image understanding, lfm3-audio with 18 language speech processing capabilities, and the general model lfm3-omni for local agents and structured output; the d1 decision model is upgraded to 2B parameters, supporting text, image, and audio decision-making, with some models supporting Apple hardware.
Ollama releasesLiquid AI has released the LFM3 small model family, including the 1.7B parameter text model lfm3 and variants supporting image and audio functions, with a context window of up to 256K tokens; the d1 decision model has been upgraded to 2B parameters, supporting text, image, and audio inference, with Apple hardware already supporting the rc1 version.
Ollama releases1515Ollama v0.40.12-rc0 released, retains small image details through MLX technology to optimize the compactness of Gemma 4 visual tower, improves prompt speed, and adds DFlash2 support.
Ollama releases3535Ollama v0.40.9 released, running on Linux and Windows platforms, improves streaming processing failure reports for OpenAI and Anthropic messages, passes cloud model context length and capabilities to coding tools via ollama launch, optimizes thinking settings for Anthropic-compatible search to avoid failures.
NVIDIA BlogNVIDIA showcased its full lineup of local AI hardware at GTC Berlin, with RTX Spark now available to support personal agents on Windows PCs, DGX Spark enabling developers to build and scale Linux applications, and DGX Station designed for Linux enterprise teams to run cutting-edge workloads, with the Windows version launching in December.
