Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Augment Code / Augment private models

Prism · Claude + GPT

Managed model ensemble that routes each request to a selected provider model.

Language models (LLMs)
availableClosed-source managed componentCodingText generation
Official documentation (opens in a new tab)Reviewed 15 Oct 2026
4 of 16 specification fields collected
Context windowNot collectedAwaiting source review
Maximum outputNot collectedAwaiting source review
Input billingSee ratesAugment · Prism routing
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected

Exact version-specific value or customer API contract not established.

Maximum input tokens
Not collected

Exact version-specific value or customer API contract not established.

Maximum output
Not collected

Exact version-specific value or customer API contract not established.

Parameters
Not collected

Exact version-specific value or customer API contract not established.

Release date
Not collected

Exact version-specific value or customer API contract not established.

Knowledge cutoff
Not collected

Exact version-specific value or customer API contract not established.

Reasoning
Not collected

Exact version-specific value or customer API contract not established.

Tool calling
Not collected

Exact version-specific value or customer API contract not established.

Structured output
Not collected

Exact version-specific value or customer API contract not established.

Streaming
Not collected

Exact version-specific value or customer API contract not established.

Batch processing
Not collected

Exact version-specific value or customer API contract not established.

Fine-tuning
Not collected

Exact version-specific value or customer API contract not established.

Context caching
SupportedSource for Context caching (opens in a new tab)

Automatic prompt caching at the managed workflow level; actual cache charges and eligibility depend on the chosen route.

Input modalities
TextSource for Input modalities (opens in a new tab)
Output modalities
TextSource for Output modalities (opens in a new tab)
Usage metering
Chosen-route token charges plus a 40% service fee on LLM usage; Cosmos compute is $0.19/hour in rounded-up five-minute increments.Source for Usage metering (opens in a new tab)

No fixed ensemble input/output tariff or universal question cost exists. Compute has no 40% service fee.

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Augment · Prism routing

Provider source (opens in a new tab)
API model ID
Not collected

No verified raw inference model selector. Product, experiment and SDK operation names are not interchangeable API IDs.

Access
Managed selection in Cosmos or Auggie CLI. Workspace rollouts and routing can change. Confirm the request-specific route in the dashboard.
Regions
No complete model-specific region inventory established.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 15 Oct 2026
CategoryUsageRateUnit
Platform service fee40% surcharge on LLM usage1$0.40USD of LLM charges
Cosmos computeSandbox compute time2$0.19hour

Rate conditions

  1. Added to underlying input, output and cache charges. No surcharge on compute. Not a token tariff or per-question price.
  2. Only when Cosmos compute is used. Billed in five-minute increments rounded up. Separate from LLM token charges and service fee.
Pricing source (opens in a new tab)

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.

Read the service policy (opens in a new tab)

Caching

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 · Prism routing

Documented
Support evidence (opens in a new tab)
How it works
Automatic caching of stable request context.Source for How it works (opens in a new tab)
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
Chosen provider-route cache-read/write rates plus the service fee on LLM charges.Source for Billing rules (opens in a new tab)
Limits & data boundaries
Model-dependent contract; no single ensemble-wide TTL, minimum or cache price established.Source for Limits & data boundaries (opens in a new tab)

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 15 Oct 2026

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

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

Augment calls Prism a proprietary ensemble. This is provider-model routing, not a privately trained base checkpoint; underlying models retain their original providers.

Classification evidence (opens in a new tab)

Sources reviewed