Configs › Kolibri-1 on DGX Spark: 32.9 tok/s measured

Kolibri-1 on DGX Spark: 32.9 tok/s measured

Measured Kolibri-1 speeds: 32.9 tok/s on DGX Spark (FP8, vllm). 1 config, each linked to its source, snapshot 2026-10-05.

Snapshot 2026-10-05 · 258 configs · 118 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.

Quick answers

How fast is Kolibri-1 on DGX Spark?
32.9 tok/s single-stream decode (FP8, vllm), the only measured config on NVIDIA DGX Spark (GB10), measured by jvr0x, source last checked 2026-10-05. source

Measured configs for Kolibri-1

ModelHardwareQuantBackendDecodeContextTrustSource
Kolibri-1DGX SparkFP8vllm32.9 tok/sStructured tablesource

More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.