Radar

Open source

Free AI models and tools that anyone can download. Newest first.

Importance 2 of 5r/LocalLLaMA

Shanghai AI Lab releases Sage S2.1 Flash scientific multimodal model

Shanghai Artificial Intelligence Laboratory published Sage S2.1 Flash on Hugging Face. The model targets multi-step research tasks, combining scientific data understanding with retrieval, programming and tool use, and supports more than ten types of scientific input.

Why it matters: Open scientific agent models give research teams a downloadable option for automating analysis workflows.

Importance 2 of 5r/LocalLLaMA

Gamer paper and models introduce a vision-language-action agent for long-horizon games

Researchers released Gamer, a generalist vision-language-action foundation model for long-horizon, real-time game agents. It combines perception, memory and multi-step reasoning and was pre-trained on tens of thousands of hours of keyboard and mouse interaction across 2D interfaces and 3D worlds.

Why it matters: Agents trained on raw device interaction point toward models that can operate graphical software and simulated environments end to end.

Importance 1 of 5r/MachineLearning

Developer builds real-time jazz partner by steering a pretrained music transformer

A developer released duetto, a real-time jazz accompanist for MIDI keyboards that plays solos, comping and walking bass. It uses the pretrained Anticipatory Music Transformer without fine-tuning, applying all control at sampling time.

Why it matters: It shows pretrained generative models can be adapted to interactive uses through sampling-time control rather than costly retraining.

Importance 2 of 5r/LocalLLaMA

Developer releases merged Qwen3.8-27B model combining vision, typed decisions and abliteration

A developer building the WebBrain agent published Qwen3.8-27B-Perplexing-Pegasus-Heretic, a single 27B model that merges vision, a fast typed-decision capability and abliterated weights. The goal is to replace two or three separate models for local web agents. It ships in GGUF, AutoRound and NVFP4 formats.

Why it matters: Consolidating agent capabilities into one local model can cut VRAM use and latency for on-premise agent setups, though abliterated weights remove safety refusals.

Only want news that affects your AI projects?
Track them in Radar and get an alert when a vendor changes something. Free to start.
Start free
Importance 2 of 5r/LocalLLaMA

Developer shows abliteration can be applied as a small runtime LoRA instead of full models

A post explains that most abliteration techniques are rank-1 ablations that fit in a tiny LoRA applied at runtime in llama.cpp. This avoids downloading separate abliterated models and can be toggled per request.

Why it matters: Refusal removal is cheap and easy to apply to open models, which matters for safety controls on self-hosted deployments.

Importance 2 of 5Simon Willison

Simon Willison releases datasette-maplibre 0.1a0, built mostly with one Claude Fable 5.2 prompt

Simon Willison released datasette-maplibre 0.1a0, a Datasette geospatial plugin combining MapLibre GL with the free OpenFreeMap vector map server. He says most of it was built from a single prompt to Claude Fable 5.2.

Why it matters: It is another practical example of frontier coding models producing usable software from a short prompt.

Importance 2 of 5r/LocalLLaMA

audio.cpp 0.9.5 adds real-time local avatar voice chat combining three open models

audio.cpp 0.9.5 adds an avatar voice chat app in its WebUI that combines SoulX-FlashHead for talking head video, OWSM v4 for speech recognition and LFM3-Audio for responses and TTS. Video generation still needs an RTX 4090 or better with cuDNN for real-time use.

Why it matters: It shows interactive voice avatars can run fully locally, though hardware demands remain high.

Importance 2 of 5r/MachineLearning

Training-free method edits one weight row so reasoning models stop thinking early

A developer showed that folding a linear probe into a single row of a reasoning model's output layer lets it end its thinking phase once it has reached an answer. The edited model needs no training or extra inference code.

Why it matters: Cutting redundant reasoning tokens could lower inference cost and latency for teams running open reasoning models.

Importance 2 of 5r/LocalLLaMA

User benchmarks 1Cat-vLLM expert caching against Strata on Tesla V100 rig

A user compared inference speeds of 1Cat-vLLM, a Tesla V100 focused fork of vLLM that recently added expert caching, with Strata. The test used a Qwen3.8-Flash-Next IQ3_XXS quant on four 16 GB V100 cards with an Epyc CPU and DDR4 memory.

Why it matters: Expert caching on older GPUs could make large mixture-of-experts models cheaper to run locally on used hardware.

Importance 2 of 5r/LocalLLaMA

VillanovaAI releases Villanova2 open model family for five European languages

VillanovaAI released the Villanova2 family for English, Italian, German, Spanish and French under Apache 2.0. It includes a 15.5B-total, 3.9B-active mixture-of-experts model and a dense 2.4B model, each in base, instruct and vision-language versions, with GGUFs available.

Why it matters: European-built open models give EU companies more options for multilingual work under a permissive license.

Importance 2 of 5r/LocalLLaMA

Open-source tool tunes cudaMemcpy to speed MoE CPU offload prefill by about 35%

A developer released a tool that tunes CUDA memory copies used when offloading mixture-of-experts layers to the CPU. On an RTX 3090 with Qwen3.6 35B-A3B in llama.cpp, prefill rose from 345 to 466 tokens per second.

Why it matters: Cheap software tuning can noticeably improve local inference on consumer GPUs for models too large for VRAM.

Importance 2 of 5r/LocalLLaMA

Basalt 0.4.0 inference engine adds Qwen 3.8 27B, reaching up to 776 tok/s on an RTX 5090

The Basalt local inference engine version 0.4.0 adds its first dense model, Qwen 3.8 27B, with up to 776 tokens per second on structured output on an RTX 5090. Recent updates also added multi-model and router support, DeepSeek V4.1 Flash, video support and files in chats.

Why it matters: Faster local inference makes running capable open models on a single workstation more practical.

Importance 3 of 5r/MachineLearning 2 sources

Marimba released as open lightweight speech decision model built on Qwen3-ASR-0.6B

A developer released Marimba, a fast, lightweight model that returns typed decisions about speech such as yes/no answers, choices or scores. It runs on a frozen Qwen3-ASR-0.6B, using zero-shot decoder scores for word and language questions and small readouts for voice activity, turn endings, speaker changes and speaker count. Weights are open under Apache-2.0 and a paper is due on arXiv.

Why it matters: Teams building voice agents could use small open components like this for turn detection and speaker handling without large models.

Importance 2 of 5r/LocalLLaMA

Axiomic Labs releases X3Guard-150M, a small prompt guard model for edge devices

Axiomic Labs released X3Guard-150M, a prompt guard model that flags whether a prompt is harmful and whether it is adversarial in a single pass. It is built on the company's GPT-X3 base model and runs in the browser with response times in the tens of milliseconds on iPhone.

Why it matters: Small guard models let teams screen prompts cheaply on device before they reach larger models.