No verified raw inference model selector. Product, experiment and SDK operation names are not interchangeable API IDs.
Access
Historical retrieval experiment; no direct vector-generation API or version pin established.
Regions
No complete model-specific region inventory established.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Augment says it does not train on customers’ proprietary code. The RLDB report distinguishes internal data and opt-in, appropriately licensed contributions. Indexes and product/account data may persist; the general privacy policy does not establish a universal fixed retention period or zero-data-retention contract.
Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUTProcess & retain
The system processes eligible input and retains reusable state.
02 / LATER REQUESTCheck available state
Reuse depends on matching input, session state and the cache policy.
03 / CACHE HITReuse & generate
Reusing state can reduce repeated work. Any charges follow the billing rules below.
Augment · internal retrieval
Not verified
No model-specific prompt-cache contract established.
How it works
Not collected
No model-specific prompt-cache contract established.
Minimum reusable input
Not collected
No model-specific prompt-cache contract established.
Lifetime & refresh
Not collected
No model-specific prompt-cache contract established.
How to enable it
Not collected
No model-specific prompt-cache contract established.
Usage fields
Not collected
No model-specific prompt-cache contract established.
Billing rules
Not collected
No model-specific prompt-cache contract established.
Limits & data boundaries
Not collected
No model-specific prompt-cache contract established.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 15 Oct 2026
What the measurements say
AugmentQA keyword recall · embeddings-only
Augment Code · provider-reported
0.74fraction
Internal repository QA evaluation. Chart gives February and March snapshots. This is provider-reported retrieval keyword recall, not independent general-purpose accuracy.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model
Why it is included
The provider explicitly describes versions of its internal trained embeddings. These dated labels identify the chart snapshots, not public checkpoint or raw embedding API names.
