> This page is for version v1.5.1.
> For other versions, use one of these documentation indexes:
> - Latest (v1.5.1) (default): https://docs.nvidia.com/dynamo/latest/llms.txt
> - dev: https://docs.nvidia.com/dynamo/dev/llms.txt
> - v1.5.1: https://docs.nvidia.com/dynamo/v1.5.1/llms.txt
> - v1.5.0: https://docs.nvidia.com/dynamo/v1.5.0/llms.txt
> - v1.4.2: https://docs.nvidia.com/dynamo/v1.4.2/llms.txt
> - v1.4.1: https://docs.nvidia.com/dynamo/v1.4.1/llms.txt
> - v1.4.0: https://docs.nvidia.com/dynamo/v1.4.0/llms.txt
> - v1.3.0: https://docs.nvidia.com/dynamo/v1.3.0/llms.txt
> - v1.2.1: https://docs.nvidia.com/dynamo/v1.2.1/llms.txt
> - v1.2.0: https://docs.nvidia.com/dynamo/v1.2.0/llms.txt
> - v1.1.1: https://docs.nvidia.com/dynamo/v1.1.1/llms.txt
> - v1.1.0: https://docs.nvidia.com/dynamo/v1.1.0/llms.txt
> - v1.0.2: https://docs.nvidia.com/dynamo/v1.0.2/llms.txt
> - v1.0.1: https://docs.nvidia.com/dynamo/v1.0.1/llms.txt
> - v1.0.0: https://docs.nvidia.com/dynamo/v1.0.0/llms.txt
> - v0.9.1: https://docs.nvidia.com/dynamo/v-0-9-1/llms.txt
> - v0.9.0: https://docs.nvidia.com/dynamo/v-0-9-0/llms.txt
> - v0.8.1: https://docs.nvidia.com/dynamo/v-0-8-1/llms.txt
> - v0.8.0: https://docs.nvidia.com/dynamo/v-0-8-0/llms.txt
> - v0.7.1: https://docs.nvidia.com/dynamo/v-0-7-1/llms.txt
> - v0.7.0: https://docs.nvidia.com/dynamo/v-0-7-0/llms.txt

> For clean Markdown content of this page, append .md to this URL. For the complete documentation index, see https://docs.nvidia.com/dynamo/llms.txt. For full content including API reference and SDK examples, see https://docs.nvidia.com/dynamo/llms-full.txt.

# Dynamo Day 0 support for TokenSpeed

> Dynamo adds day-0 TokenSpeed support with the Dynamo frontend for Kimi K2.5.

[TokenSpeed](https://lightseek.org/blog/lightseek-tokenspeed.html) ([GitHub](https://github.com/lightseekorg/tokenspeed)) launched today as LightSeek's new inference engine for agentic workloads. The initial repo is a preview, with more model coverage and runtime features landing over the next few weeks.

Two pieces are worth calling out. First, TokenSpeed includes new MLA kernel work for long-context Kimi-style workloads on Blackwell. Second, TokenSpeed has a native C++ scheduler in `tokenspeed-scheduler/` that models request flow and cache operations as explicit state machines, while Python remains the runtime and integration layer.

Dynamo now has day-0 support for running TokenSpeed as a Dynamo backend through `python -m dynamo.tokenspeed`. The Dynamo frontend remains the user-facing OpenAI-compatible API entrypoint and handles request routing, streaming responses, and cancellation.

See the [Kimi K2.5 TokenSpeed recipe](https://github.com/ai-dynamo/dynamo/tree/v1.5.1/recipes/kimi-k2.5/tokenspeed/agg/nvidia) for the current Dynamo launch recipe.

Things are moving quickly. Upstream TokenSpeed calls out ongoing work on model coverage, P/D, EPLB, KV store, Mamba cache, VLM, metrics, Hopper optimization, and related runtime features.