models · thinkingmachines
Inkling Small — catalogued from third-party evaluation data; no weights held by Sovereign Frontier.
| Method | hf-published-sha256 |
|---|
| Publisher | thinkingmachines |
|---|---|
| Origin | United States |
| Licence | Apache-2.0 |
| Parameters | 265,956,439,090 (266B) |
| Context window | 1,048,576 tokens |
| Modality | text+image+audio->text |
| Native precision | BF16 |
| Downloads reported upstream | 73,353 |
| GPQA Diamond | 89.5 |
|---|---|
| Humanity's Last Exam | 33.3 |
Reported by a third party and recorded here, not re-run by us. Source: artificialanalysis.ai/api/v2.
| Hardware | Engine | Build measured | Decode, 1 stream | Label | Reported by |
|---|---|---|---|---|---|
| RTX PRO 6000 Blackwell ×2 devices | sglang | ModelOpt NVFP4 | 115 tok/s @ 50 | candidate | ppickle1989 |
| GeForce RTX 3090 ×6 devices | llama.cpp | IQ4_XS | 47.4 tok/s @ 262,144 | candidate | localmaxxing |
| NVIDIA DGX Spark GB10 ×2 devices | vllm | ModelOpt NVFP4 | 32.3 tok/s @ 563 | candidate | 0xsero |
From local-ai-registry (MIT), commit 124959d: 4 runs across 3 machines, 0 validated — model revision and runtime pinned, launch accepted — and 4 candidate, their label for useful evidence without a reproducible-launch promise.
Registry record as of 2026-08-15. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.