opensource/coder
3
0
Fork 0
mirror of https://github.com/coder/coder.git synced 2026-09-24 15:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: price AI usage by configured provider type (#27836)

Browse Source 
## Problem

AI Gateway records the aibridge provider on each interception, which is
the upstream wire format and only ever `anthropic`, `openai`, or
`copilot`. Prices are matched on exact provider and model equality, so a
provider configured as Azure, Bedrock, Google, OpenRouter, or Vercel is
priced as if it were native OpenAI or Anthropic, matching either the
wrong price or no price at all.

## Changes

- Resolve the configured provider type from `ai_providers` by provider
name, which is unique among live providers, and key the price lookup on
it instead of the aibridge provider. No schema change is needed.
- Label `unpriced_token_usage_records_total` with the same provider
value used for the lookup, so it names a provider an operator actually
configured.
- Treat a provider that cannot be resolved as unpriced, consistent with
how a missing price is handled today.

Closes
https://linear.app/codercom/issue/AIGOV-570/resolve-ai-model-prices-using-the-configured-provider-type
Depends on the follow-up that extends the shipped price book to the
remaining provider types:
https://linear.app/codercom/issue/AIGOV-571/ship-prices-for-all-ai-governance-provider-types

> [!NOTE]
> Initially generated by Claude Opus 5, modified and reviewed by
@ssncferreira
This commit is contained in:
Susana Ferreira
2026-08-04 14:44:27 +01:00
committed by GitHub
parent 4aec1ea592
commit db88ec3f6a
6 changed files with 357 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/admin/integrations/prometheus.md
+1 -1
View File
@@ -125,7 +125,7 @@ The `coder_ai_gateway_cost_control_*` metrics are exported only by `coderd`.
| `coder_ai_gateway_cost_control_blocked_requests_total` | counter | The number of AI requests blocked because the initiator's budget was exceeded. | `group_id` |
| `coder_ai_gateway_cost_control_blocked_users` | gauge | The number of users currently over their AI budget. | `group_id` |
| `coder_ai_gateway_cost_control_enforcement_duration_seconds` | histogram | The duration of AI budget enforcement checks, in seconds (outcome: allowed, blocked, error). | `outcome` |
| `coder_ai_gateway_cost_control_unpriced_token_usage_records_total` | counter | The number of recorded AI token-usage records for which no model price was found (provider: anthropic, openai, copilot). | `model` `provider` |
| `coder_ai_gateway_cost_control_unpriced_token_usage_records_total` | counter | The number of recorded AI token-usage records for which no (provider, model) price was found. | `model` `provider` |
| `coder_ai_gateway_injected_tool_invocations_total` | counter | The number of times an injected MCP tool was invoked by AI Gateway. | `model` `name` `provider` `server` |
| `coder_ai_gateway_interceptions_duration_seconds` | histogram | The total duration of intercepted requests, in seconds. The majority of this time will be the upstream processing of the request. AI Gateway has no control over upstream processing time, so it's just an illustrative metric. | `model` `provider` |
| `coder_ai_gateway_interceptions_inflight` | gauge | The number of intercepted requests which are being processed. | `model` `provider` `route` |