---
type: "raw_signal"
id: "how-to-serve-trillions-of-tokens-for-trillion-894224"
slug: "how-to-serve-trillions-of-tokens-for-trillion-894224"
title: "How to serve trillions of tokens for trillion"
description: "Modal published a production engineering account of serving coding-agent workloads with a roughly trillion-parameter Kimi model."
retrieval_nugget: "Modal published a production engineering account of serving coding-agent workloads with a roughly trillion-parameter Kimi model."
observed_at: "2026-09-23"
record_date: "2026-09-23"
date_kind: "observed_at"
why_it_matters: "Source: https://modal.com/blog/trillion-tokens-trillion-parameters"
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "modal.com"
topics: []
entities: []
related_patterns: []
source_url: "https://modal.com/blog/trillion-tokens-trillion-parameters"
source_urls: ["https://modal.com/blog/trillion-tokens-trillion-parameters"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:how-to-serve-trillions-of-tokens-for-trillion-894224"
status: "published"
visuals: []
routes: {"html":"https://newruntime.com/signals/how-to-serve-trillions-of-tokens-for-trillion-894224/","markdown":"https://newruntime.com/signals/how-to-serve-trillions-of-tokens-for-trillion-894224.md","json":"https://newruntime.com/signals/how-to-serve-trillions-of-tokens-for-trillion-894224.json"}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
---

# How to serve trillions of tokens for trillion

## Retrieval answer

Modal published a production engineering account of serving coding-agent workloads with a roughly trillion-parameter Kimi model.

Modal published a production engineering account of serving coding-agent workloads with a roughly trillion-parameter Kimi model. Its representative workload had about 100,000 input tokens and 500 output tokens per request, while repeated session history made KV-cache management central. Modal reports 2.8× higher per-user replica performance and 5.6× higher aggregate performance across users after serving optimizations built on SGLang.

Source: https://modal.com/blog/trillion-tokens-trillion-parameters

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)