AI Briefing#qwen
The latest AI and developer news about #qwen, with the original source and a short summary.
The latest AI and developer news about #qwen, with the original source and a short summary.
A 32B medical LLM built with SeedRL was released, with thinking mode set as the default.
With DFlash in oMLX 0.3.5 RC1, Qwen3.5 27B went from 9 to 22 T/S on a Mac M5 Max.
A coding agent model, Qwen3.6-35B-A3B, has been released for anyone to use.
Shared real-world benchmarks and issues from running Qwen3.5-27B on the Arc B70.
Running MiniMax M2.7 AWQ on 2 GX10s to complete a local agentic coding stack.
Frontier open models require sustained funding, so they will likely end up moving toward a consortium.
KV cache compression enables memory sharing across multi-agents, cutting tokens by up to 49%.
In an OpenClaw+Ollama setup, glm-4.7 was more stable for agentic tasks than Qwen or Gemma.
Built a free web search and scraping workflow using local Qwen3.5 and MCP tools.
Model racing techniques cut RAG query rewriting latency by more than 50% while securing system resilience.
For 96GB VRAM setups, Qwen 3.5 122B and 27B have emerged as the practical choices.
A practical build and optimization advice for creating an offline companion robot with an 8GB ThinkPad was shared.
Verified Qwen3.5-122B running at 198 tok/s on 2x RTX PRO 6000 Blackwell.
Sentence Transformers v5.4 lets you compare text, image, audio, and video with the same API.
A self-distillation (SSD) technique that uses only the model's own outputs, without a separate teacher model or reinforcement learning, significantly improved LLMs' code generation performance.
Alibaba has unveiled Qwen3.6-Plus, which significantly enhances coding and multimodal agent performance.
Alibaba's Qwen 3.6 Plus is now available in Vercel AI Gateway.
Qwen3.5-Omni scales up model size with the goal of native omni-modal AGI.
Using AutoBe and Typia, a deterministic feedback loop overcomes low Function Calling success rates and pushes them close to 100%.
Qwen3.5-Max-Preview has been deployed on Arena and shown strong performance in preliminary evaluations.
By attaching CPT and BM25 RAG to Qwen3-30B-A3B, the team won the Excellence Award while running in a 24GB environment.
It automatically finds behavioral differences between models, surfacing hidden risk signals.