vllm-sr open-sources d3 reasoning model with xhigh/medium/low thinking levels on HF
victormustar · x · 2026-10-12
Developer victormustar announced the release of vllm-sr/d3 on Hugging Face. Its chat template reveals an open-source multimodal reasoning model (image/video input) with a built-in reasoningeffort mechanism — defaulting to xhigh with instructions to validate assumptions and weigh alternatives, plus medium/low tiers for cheaper inference. The template handles system messages, video counting, and vision ID injection, ready for vLLM deployment.
More from Models
- ChatGPT glitches on iPhone 15 Pro whenever it renders image comparisons — adam0672 · 2026-10-12
- Claude reportedly planned and coded a full video production in 2.5 hours — ZeroStateReflex · 2026-10-12
- KV caching explained: why it shapes GPU memory and drove the shift to GQA and MLA — blaizedsouza · 2026-10-12
- Claude Opus 5.5 vs GPT-6.1 Sol hands-on: this time GPT wins the vibe-check — CodeByPoonam · 2026-10-12
- DeepMind's AlphaProof Nexus reportedly cracks 9 Erdős problems, 44 OEIS conjectures — kimmonismus · 2026-10-12
- Codex keeps stalling on legit debugging over cybersecurity guardrails, user complains — Al_Grigor · 2026-10-12