Thinking Machines open-sources 500B MoE Inkling 2 Small under Apache 2.0, vLLM ships Day-0 support
woosuk_k · x · 2026-11-13
Thinking Machines Lab has released Inkling-2-Small, with weights downloadable on HuggingFace under an Apache 2.0 license.
vLLM announced Day-0 support: a 500B-parameter MoE model with text + image inputs and text output, letting you bring images into the conversation on the same model. Under the hood: a new ViT image encoder, mixed NVFP4/MXFP8 MoE layers, and DSpark speculative decoding.
The vLLM integration, led by Inferact in collaboration with Thinking Machines, spans MoE and attention kernels, the vision encoder and image preprocessing, and draft-model GEMMs for speculative decoding.
Related event: Thinking Machines Open-Sources Inkling 2 Small Weights Under Apache 2.0(20 posts)→
More from Infra
- Batch APIs: the 50% LLM discount only pays off after model choice and prompt caching — arpit_bhayani · 2026-11-13
- RTX Pro 6000 bricks mid-Qwen run, owner jokes it "died of a broken heart" over Jensen news — wine_dark · 2026-11-13
- Bittensor's Chutes Handles 256 Concurrent Requests on a Single RTX 5090 at ~18k tok/s — markjeffrey · 2026-11-13
- Google expands TPU v8 lineup: 8tx cuts latency 2x, 8i Max doubles silicon per rack — Beth_Kindig · 2026-11-13
- Horace He to share two mental models for reasoning about inference at SGLang Summit — cHHillee · 2026-11-13
- Used A100 prices fall another 20-30% as sellers rush ahead of a collapse — Bascilian · 2026-11-13