AI Radar

DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models

Online reinforcement learning (RL) fine-tunes diffusion models with black-box rewards, but reverse-process methods such as Flow-GRPO are limited in efficiency by poor credit assignment, while forward-process methods such as DiffusionNFT are efficient yet unstable without carefully tuned exponential-moving-average (EMA) off-policy training. We show that NFT-style objectives contain a hidden forward--reverse consistency force that joint positive--negative training cancels only partially, and that EMA stabilizes training through restoration rather than off-policy rollouts. We introduce…

Same scene, 30 art styles, 8 local models on a Mac (mflux): Qwen-Image 2.1, FLUX.2 Klein, Krea 2, Z-Image, Ming-Image, FIBO 1.5, ERNIE, Boogu

I got into text-to-image when Qwen-Image 2.1 came out, and started playing with it on my Mac with mflux 0.22. I was impressed by how many art styles it could pull off, so I ran one simple prompt through it with only the art style changed, 30 times. Then I got curious how the other models would handle the same 30 styles. About a day and a half later, on and off (rendering, fixing prompts, re-rendering), here it is: link A row per style, a column per model, with the prompt beside each row. Tap a picture to open it, then swipe (or use the arrow keys) across models or down through styles. I like…

Meet Cloudflare Clef-omni: A 30B-A3B Multimodal Decision Model for Audio, Video, Image and Text

Cloudflare released Clef-omni, an Apache-2.0 multimodal decision model built on Qwen3-Omni-30B-A3B. It scores text, images, audio and video in one call for $0.15 per million input tokens, while Clef-flash drops to $0.038. The post Meet Cloudflare Clef-omni: A 30B-A3B Multimodal Decision Model for Audio, Video, Image and Text appeared first on MarkTechPost .

Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition

Adversarial games have driven advances from heuristic search to reinforcement learning, yet learning and adapting strategies from limited samples remain challenging. AI agents offer an alternative by turning game experience into revisions of executable policies. Building on heuristic learning (HL), we formalize Adversarial Heuristic Learning (AHL), a paradigm that uses AI agents as learning engines to refine game policies and supporting software while keeping model weights fixed. We introduce AAArena, a benchmark comprising 12 authentic adversarial games and 1,920 archived human programs…

The gap in shared understanding of LLM capability is widening

Asana cuts model costs 76x in browser tests with GPT-6.1 Sol

Using GPT-6 Astra in Codex, Asana made its browser agent 76x cheaper and 5x faster in tests to offer customers more capable models.

Qwen released QwenGyre - RL framework for training AI agents on extremely long complex tasks

Introducing QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. It launches with two core designs: - Elastic Scheduler: reallocates GPUs between rollout & training as demand shifts—live executions never interrupted. Up to 1.85× speedup over Colocate, 1.78× over Async. - Trajectory Processor: turns branching executions (compaction, sub-agents, retries) into prefix-shared trajectory trees with bounded, deduplicated training samples. Partial progress after timeout still gets scored. Scaled to Qwen3.8-2.4T-A95B with ~700K tokens per rollout, QwenGyre lifts NL2RepoBench…

Boole-1.7B: A 1-bit Qwen3-1.7B model that outperforms Bonsai-1.7B

We're launching Precision Numerics by releasing Boole-1.7B, a 1-bit version of Qwen3-1.7B. 247 MB packed weights: 13.9× smaller than BF16. 85.9% of the BF16 model’s average score across our 10-task evaluation - outperforms 1-bit Bonsai-1.7B and nearly matches 1.58-bit Bonsai-1.7B. Try it locally in your browser (works best with English), or use it with vLLM or llama.cpp. We'll ship more capable models for the open-source community. Let us know what you what models/capabilities you want to see next. Blog: precisionnumerics.ai/blog/boole-1p7b Weights…