thinkingmachines models
2 thinkingmachines models from thinkingmachines on Hugging Face, from 266.0B to 952.4B parameters, published by thinkingmachines in BF16, with a 1M-token context. The smallest official model, Inkling-Small, needs about 594 GB of VRAM at its published precision; the cheapest live fit is A100 at $9.41/hr.
Pick a size
One row per official thinkingmachines size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| Inkling-Small | 266.0B | 594 GB | 297 GB | 149 GB | A100 × 8 | $9.41/hr | 5.3 GB |
| Inkling | 952.4B | 2129 GB | 1064 GB | 532 GB | No live fit | – | 8.3 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Inkling-Small | 266.0B | BF16 | 594 GB | A100 × 8 | $9.41/hr |
| Inkling | 952.4B | BF16 | 2129 GB | No live fit | – |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.