Mastodawn

Ivan Fioravanti ᯅ (@ivanfioravanti)

vllm-metal에서 PR 작업을 통해 M3 Ultra에서 최대 32K 컨텍스트로 Qwen3-0.6B 테스트를 실행한 내용이 공유됐다. mlx-lm이 여전히 더 낫다고 언급하며, TTFT는 적절한 지표가 아니고 개선 작업이 진행 중이라고 밝혔다. 히트맵과 Prefill/Decode 차트도 업데이트됐다.

https://x.com/ivanfioravanti/status/2036156678035419263

#vllm #mlxlm #qwen3 #llm #opensource

Ivan Fioravanti ᯅ (@ivanfioravanti) on X

I did my first PR on vllm-metal and I was able to run tests with up to 32K context on M3 Ultra with Qwen3-0.6B. mlx-lm is still the winner and TTFT is not the correct one there, improvement is WIP. Mega heatmap andPrefill/Decode charts updated. ctx rows leverages caching.

X (formerly Twitter)

Dr. Simonas Kareiva LY2EN 2d ago

Running a local #qwen3 model over #opencode produced the below vibe coding results - an Ansible reference. It's not much but it was honest work :) https://git.wtf.lt/simonas/ansible/src/branch/master/README.md

ansible/README.md at master

ansible

Gitea: Git with a cup of tea

Techgedöns 👨🏼‍💻📱 #Tchgdns 2d ago

Petal für #macOS: Simple Diktat-App für die Menüleiste mit #Qwen3, #Parakeet und #Voxtral. #OpenSource, ohne Cloud und Abo 🎙️🖥️🔓 https://tchgdns.de/?p=162724

Hacker News 5d ago

MacBook M5 Pro and Qwen3.5 = Local AI Security System

https://www.sharpai.org/benchmark/

#HackerNews #MacBook #M5 #Pro #Qwen3.5 #Local #AI #Security #System #AI #Security #Technology #News

HomeSec-Bench — Local AI vs Cloud Benchmark | SharpAI Aegis

Qwen3.5-9B scores 93.8% on 96 real security AI tests — within 4 points of GPT-5.4 — running entirely on Apple Silicon. Full benchmark results and methodology.

openSUSE Linux 6d ago

This week on #openSUSE Planet, find out about a new #Cockpit launcher that simplifies #sysadmin tasks, Cavil #Qwen3.5-4B brings #AI-powered #legal compliance, and #LogAI lets you query system logs for overnight answers. #Linux https://news.opensuse.org/2026/03/20/planet-roundup/

Planet News Roundup

This is a roundup of articles from the openSUSE community listed on planet.opensuse.org. The community blog feed aggregator lists the featured highlights bel...

openSUSE News

sayzard Mar 18

Sebastian Raschka (@rasbt)

Mamba-3이 출시되었으며, 작성자는 Mamba 및 유사 모델들이 트랜스포머 어텐션 하이브리드 아키텍처(Qwen3.5, Kimi Linear 등)에서 흥미로운 활용처라고 평가합니다. 다음 세대 하이브리드에서 Gated DeltaNet 대신 RoPE가 추가된 Mamba-3을 교체해보는 실험을 제안하고 있습니다.

https://x.com/rasbt/status/2034088726997893168

#mamba3 #transformer #qwen3.5 #gateddeltanet #rope

Sebastian Raschka (@rasbt) on X

Oh wow, Mamba-3 is here! For me, the most interesting use case of Mamba and Mamba-likes are the recent transformer attention hybrid architectures (Qwen3.5, Kimi Linear, etc.) Would be interesting to swap Gated DeltaNet with Mamba-3 (which now also has RoPE) in next gen hybrids.

X (formerly Twitter)

sayzard Mar 15

Prince Canuma (@Prince_Canuma)

mlx-embeddings v0.1.0 출시: 새 모델로 Alibaba의 Qwen3 VL Embedding 및 Reranker와 ColDefics3(LoRA 어댑터 및 ColVision 프로세서 포함)이 추가되었습니다. NVFP4·MXFP4·MXFP8 양자화 지원이 도입되었고, Gemma3의 양방향 모델 임베딩 품질 수정이 포함됩니다.

https://x.com/Prince_Canuma/status/2032890809847029896

#mlxembeddings #embeddings #qwen3 #coldefics3 #quantization

Prince Canuma (@Prince_Canuma) on X

mlx-embeddings v0.1.0 is out! 🔥 New models: → Qwen3 VL Embedding and Reranker by @Alibaba_Qwen → ColDefics3 with LoRA adapters & ColVision processor b New features: → NVFP4, MXFP4 and MXFP8 quantization support → Gemma3 embedding quality fix for bidirectional models →

X (formerly Twitter)

Nicolas MØUART Mar 15

That's an excerpt of #Qwen3.5 35B summary on Robin Wright after I made it correct everything :
"Robin Wright is an acclaimed American actress, producer, and director. Born in 1966, she gained early fame as Princess Buttercup in The Princess Bride (1987) and earned a Golden Globe nomination for Forrest Gump (1994). She achieved global acclaim as Claire Underwood in the #Netflix series House of Cards (2013–2018), winning a Golden Globe ..""
Not a word on Santa Barbara #TV show 🤷‍♂️
#80s