vllm-sr/d3-ultra
just landed · new, unvettedmodel
New, unvetted: created or first seen in the last few days, not from a trusted org and not past the signal threshold yet.
d3-ultra is the newest 27B multimodal decision model of Decision 3.0, the decision models of vLLM Semantic Router. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0.
- Org
- vllm-sr
- Released
- 2026-10-24
- Status
- just landed · new, unvetted
- License
- apache-2.0 (permissive)
- Parameters
- 26.1B
- Architecture
- dense · Qwen3_5Model
- Context
- 262,144 tokens
- Task
- zero-shot-classification · other
- Signal
- 14 HF likes · 56 downloads
Runs on
mixed: real files + estimate; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~18.8 GB (estimated size): 24GB GPU (3090 / 4090), 24GB GPU (7900 XTX), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~30.8 GB (estimated size): 32GB GPU (5090), 48GB GPU (RTX 6000 Ada / A6000), Arc B70 32GB, Mac 48GB, Mac 64GB (M1 Max)
- FP16 ~55.3 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 64GB RAM (slow), CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)
Links
First published here 2026-10-24 · updated 2026-10-26 · markdown · JSON