OpenCode Cache Hit Rate with Claude Models

Key takeaways

• What is normal: 96.5%, median of 7 users in running conversations; middle half 94.4%–97.5%
• All requests: 90.3%, including routing, expiry and new chats
• Largest cause of misses: account switching, 37% of misses

What cache hit rate is normal for each client

Week of 2026-09-29 – 2026-10-05 (UTC)

• Continuation hit rate (median user): 96.5%
• Middle half of users: 94.4%–97.5%
• Qualifying users: 7 (at least 20 continuation turns in the week)
• Hit rate over all requests: 90.3%

Why is my OpenCode cache hit rate low? Misses by cause

9.7% of this week's input was not read from cache. Here it is by cause, with what you can do about each.

Routed to another upstream account or endpoint: 37% of misses, about 3.5% of all input

A prompt cache lives with the upstream account that served the request. When a relay or multi-account gateway sends the next turn to another account, the whole context is written again. Direct connections to the vendor never show this cause.

New content and prefix changes in a running conversation: 31% of misses, about 3.0% of all input

Content that is new in a turn can only be written, not read, which is unavoidable; a changed system prompt, tool list or rewritten history also invalidates the prefix.

More than 5 minutes idle, cache expired: 19% of misses, about 1.8% of all input

More than 5 minutes passed between requests, so the previous cache entry may have expired.

New conversation, or idle for over an hour: 13% of misses, about 1.3% of all input

The first request of a new conversation, or one after an hour of idling, has to write the cache again.

Parallel request (previous one still running): 1% of misses, about 0.1% of all input

A request sent before the previous one finished (subagents, parallel tool calls) cannot read the cache the previous one is still writing.

Switched model: 0% of misses, about 0.0% of all input

Each model has its own cache, so the first request after a model switch reads the whole history again.

A few points of hit rate, several times the bill

At Claude Opus 5.5 official prices, with misses written to the 5-minute cache, the effective cost per million input tokens:

• This week's median, 96.5%: $0.37 (×1.0)
• 90%: $0.68 (×1.9)
• 80%: $1.16 (×3.2)
• 50%: $2.60 (×7.1)

Run your own numbers in the Claude API cost calculator

How to check your OpenCode cache hit rate

• Anthropic format: cache_read_input_tokens ÷ (input_tokens + cache_read_input_tokens + cache_creation_input_tokens).
• OpenAI format: cached_tokens (in prompt_tokens_details or input_tokens_details) ÷ prompt_tokens / input_tokens.
• On Wokey, every request in the console's request log shows its cache reads and cache writes.

Why hit rates drop behind relays

• Account switching: the cache belongs to the upstream account, so a turn sent to another account rewrites everything. The most common relay cause; direct connections never have it.
• Format conversion: translating between Anthropic and OpenAI formats can change the prefix or drop cache_control breakpoints.
• Claude Code before v2.1.181, on a custom base URL, put a per-request token in the attribution block at the top of the system prompt, which defeats gateways that cache on the request body; upgrade, or set CLAUDE_CODE_ATTRIBUTION_HEADER=0.
• Self-test: send two requests with the same prefix above the minimum cache length; the second one's cache reads should be close to the prefix length.

Wokey routes a conversation's later requests to the same upstream account whenever it can. From 2026-10-04 to 2026-10-10, ordinary Wokey users hit 92.8% on Claude Opus 5.5 over all requests, against 87.8% for OpenRouter overall.

Compare with OpenRouter cache hit rates · Live: Wokey public cache hit-rate API (last 7 days)

How we measure

• Real user requests through the Wokey gateway; every Monday at 05:00 UTC we compute the previous Monday-to-Sunday week.
• Clients are identified from the request User-Agent. Requests from SDKs, HTTP libraries and downstream relays, whose real client is unknown, are left out, as are enterprise accounts.
• A continuation turn has the same model and upstream account as the previous request of its conversation and starts within 5 minutes of it finishing. Each user's rate is cache reads ÷ input over those turns (input includes cache reads and writes); we report the median over users with at least 20 such turns in the week.
• A cell is published with at least 5 qualifying users; a change of more than 3 points from last week is checked by hand first.
• Known biases: Wokey users only; usage habits vary widely; the all-requests rate includes the gateway's own routing loss (the "routed to another upstream account" cause).

FAQ