Prompt Caching

Prefix caching is on for every open source model. No configuration, no cache-write surcharge. When a request shares a prefix with earlier traffic (system prompt, tool definitions, conversation history), those tokens skip prefill and bill at the cached rate.

Model	ID	Input per 1M	Cached input per 1M	Output per 1M
GLM-5.2 744B	`morph-glm52-744b`	$1.10	$0.22	$4.10
Kimi K3 ^{coming soon}	—	—	—	—

Cached input on GLM-5.2 is 80% off. Other open source models cache automatically too; their cached tokens currently bill at the regular input rate.

Reading cache hits

Every response reports how much of the prompt was served from cache:

{
  "usage": {
    "prompt_tokens": 18211,
    "completion_tokens": 512,
    "total_tokens": 18723,
    "prompt_tokens_details": { "cached_tokens": 17408 }
  }
}

cached_tokens billed at the cached rate, the remainder of prompt_tokens at the input rate.

Getting hits

Prompt caching: a request matching the stable prefix is a cache hit and reuses those tokens; changing any token in the prefix is a cache miss

Matching is exact-prefix and block-aligned. To maximize hit rate:

Put stable content first: system prompt, then tool definitions, then history. Variable content (the user’s latest message, retrieved context) goes last.
Keep the prefix byte-identical between turns. A timestamp or request ID in the system prompt kills every hit after it.
Short prompts rarely hit. Caching operates on ~1k-token blocks, so a 300-token prompt has nothing to reuse.

Multi-turn agent loops get this for free: each turn re-sends the previous turns verbatim, so everything but the newest turn is a cache hit.

Cache TTL

By default cached prefixes persist under LRU eviction, with no fixed expiry. To control retention per request, pass cache_ttl:

Value	Retention
`"5m"`	5 minutes
`"30m"`	30 minutes
`"1h"`	1 hour
`"6h"`	6 hours
`"24h"`	24 hours

Expiry is sliding, Anthropic-style: every cache hit refreshes the clock. Past the TTL the prefix stops hitting entirely (full recompute), and re-sending it caches it fresh. A prefix shared by multiple requests keeps the longest surviving TTL.

cache_ttl is rolling out now, GLM-5.2 first and Kimi K3 at launch. Requests that include it are accepted today; the field takes effect as each model’s rollout completes.

cURL
Python
TypeScript

curl -X POST "https://api.morphllm.com/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "morph-glm52-744b",
    "messages": [{"role": "user", "content": "..."}],
    "cache_ttl": "1h"
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.morphllm.com/v1",
)

# Non-standard fields go via extra_body
response = client.chat.completions.create(
    model="morph-glm52-744b",
    messages=[{"role": "user", "content": "..."}],
    extra_body={"cache_ttl": "1h"},
)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://api.morphllm.com/v1",
});

const response = await client.chat.completions.create({
  model: "morph-glm52-744b",
  messages: [{ role: "user", content: "..." }],
  // @ts-expect-error non-standard field
  cache_ttl: "1h",
});

Fine print:

Invalid cache_ttl values are rejected with a 400. Only the five tiers above are accepted.
Expiry granularity is ~30 seconds: treat a TTL as “at least this long, expired within ~30s after.”
Omitting cache_ttl keeps the default behavior (LRU, no fixed expiry).

Pitfalls

cached_tokens is 0 on every request

Your prefix is changing between requests. Diff two consecutive prompts byte-for-byte; the first divergent token ends the cacheable prefix. Common culprits: timestamps, UUIDs, or shuffled tool order in the system prompt.

Hits stop after an edit mid-conversation

Editing an earlier message invalidates everything after it. Expected: caching is prefix-based, so append, don’t rewrite.

Get Started

Models

API Reference

Integrations

Reading cache hits

Getting hits

Cache TTL

Pitfalls

See Also

​Reading cache hits

​Getting hits

​Cache TTL

​Pitfalls

​See Also

Reading cache hits

Getting hits

Cache TTL

Pitfalls

See Also