# Cache warming The `warm` managed extension repeats a session's last sent provider request while the idle session waits for work. It uses a reduced output budget and a background call. This spends provider usage and does not guarantee a cache hit. [`warm/extension.gleam`](../src/albedo/harness/extensions/warm/extension.gleam) owns the worker and its scheduling decisions. ## Eligibility and settings `wanted` asks `Session.awaited()` whether work that can wake the session is running. The captured call also needs a usable TTL and enough cached input. The count is cached input tokens plus cache creation tokens. Missing usage counts as zero. `warm.minCachedTokens` defaults to 1024 and is read when planning a ping. See [`warm/extension.gleam`](../src/albedo/harness/extensions/warm/extension.gleam) and the [extension architecture](extensions.md). A request with budgeted thinking cannot be warmed: Anthropic keeps the thinking budget below `max_tokens`, so a one-token ping cannot carry it, and dropping the thinking would change the cached prefix. The `anthropic-api` provider refuses such a ping before sending it. The refusal ends warming for that idle stretch and shows as one failed background request. Adaptive thinking has no budget and warms like any other request. Explicit cache marks supply their shortest TTL. Without marks, the first clock tier of a matching `Refresh` or `Fixed` entry supplies it. See the [TTL table](cache-ttl.md) for matching and layer precedence. ## Scheduling and stopping A successful `CallSent` replaces the stored call and resets the ping count. An uncancelled `TurnEnded` schedules a ping if the plan and awaited work permit it. `Stirred` invalidates older scheduled ticks, and `Compacted` clears the call. A restored call can schedule a ping when the worker has no call already. [`step` and `due`](../src/albedo/harness/extensions/warm/extension.gleam) implement these transitions. For TTLs over ten seconds, the interval is the smaller of 90% of the TTL and the TTL minus ten seconds. Shorter TTLs use 90%. The clock starts at request start or response finish according to the table entry. A tick after `latest_ms` is discarded. The cap is `max(0, floor(write / read) - 1)` pings, using default multipliers of 1.0 for writes and 0.1 for reads when necessary. A failed call or a ping without cached input after a cached original call ends the sequence. These are estimates based on table inputs, not a billing guarantee. See `plan`, `resolution`, `interval_ms`, `cap`, and `cache_lost` in the [warmer source](../src/albedo/harness/extensions/warm/extension.gleam). ## Background requests and restoration `ping` copies the captured request, sets `patient` to false, and sets `max_output_tokens` to 1 for Chat Completions or 16 for Responses. It sends the original prefix identity through `Session.call`. A refused or failed call does not schedule another ping. See [`warm/extension.gleam`](../src/albedo/harness/extensions/warm/extension.gleam) and [provider requests](provider-requests.md). The worker keeps its captured call in memory. [`session_last_call.gleam`](../src/albedo/daemon/session_last_call.gleam) rebuilds a candidate after restart and checks its identity before restoration. A changed request can therefore prevent restoration rather than warming a different prefix. ## Related pages - [All topics](README.md) - [Change guidance](contributing.md) - [Prompt cache TTL table](cache-ttl.md) - [Durable operation admission](operations.md)