276B Param MoE Model Runs on Single B300, vLLM Creator Praises Inkling-Small
woosuk_k · x · 2026-07-31
Woosuk Kwon, creator of vLLM, reshared and praised the newly released Inkling-Small by Thinkymachines, noting that it is small enough to run on a single NVIDIA B300 GPU and might be the best model in its size class.
According to the quoted vLLM announcement, Inkling-Small features 276B total parameters with 12B active. It natively supports text, image, and audio inputs, and boasts a 1M-token context window. The open-weight model also ships with Day 0 vLLM support optimized for the Blackwell architecture.
Related event: Inkling-Small: 276B params, 12B active, open-source tops ARC(48 posts)→
More from Infra
- NVIDIA launches RTX Spark for local AI on Windows PCs — ostrisai · 2026-10-16
- SkyRL ships Day-0 RL support for Beam with +62% LoRA decode throughput in vLLM — vllm_project · 2026-10-16
- DFlash2 delivers up to 3.03x decode-only speedup for Beam across five benchmarks — vllm_project · 2026-10-16
- vLLM's Beam serving kernels cut GB200 forward time from 8.15 to 6.47 ms at TP8 batch 128 — vllm_project · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16