// Thinking Machines Inkling · Vision-language
Inkling-Small: VRAM & local deployment
276B total · 12B active (MoE) parameters. About 185 GB at 4-bit and 608 GB at BF16.
released 2026-07-27licence: Commercial OKmemory figures are estimates
01Specs
| Publisher | Thinking Machines Lab · thinkingmachines |
|---|---|
| Family / series | Thinking Machines Inkling · Inkling-Small |
| Type | Vision-language (image-text-to-text) |
| Parameters | 276B total · 12B active (MoE) Source: official model card“276B total, 12B active” |
| Context length | — |
| Architecture | 42 layers · GQA · MoE |
| Released weights | BF16 · 532 GBdownload |
| Licence | Apache-2.0 · Commercial OKPermissive licence (Apache-2.0 / MIT / CC-BY / OpenMDW): commercial use allowed. |
| Released | 2026-07-27 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.
| Precision | Weights | Smallest GPU | Mac |
|---|---|---|---|
| BF16 / FP16 (original) | 552 GB | 8 × H200 (1,128GB) | — |
| FP8 / INT8 / Q8_0 | 293 GB | 4 × RTX PRO 6000 (384GB) | Mac 512GB |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 167 GB | 4 × RTX PRO 6000 (384GB) | Mac 512GB |
KV cache not estimated: this model uses a new compressed / sparse attention design that our formula does not cover. Totals above are weights + overhead only.
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- vLLM — mentioned in the official model card
- SGLang — mentioned in the official model card
- Transformers — mentioned in the official model card
04Official repositories & quantised builds
2 official repo(s) from thinkingmachines.
- Inkling-Small main
- Inkling-Small-NVFP4 NVFP4
05Other Thinking Machines Inkling models
All 2 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| Inkling | 975B / 41B | 651 GB | 2026-07-14 |
06Sources
Data checked 2026-07-31. Parameters, context and licence are copied from the official repository; memory figures are our estimates.