fix(opencode-review): replace free OpenRouter pins with paid coder models#622
Merged
Conversation
…dels Free-tier candidates hit the shared free-models-per-day cap (free-models-per-day-high-balance) and hung for the full 5400s candidate timeout each, exhausting the pool. Live probe confirmed every :free model 429s once the daily cap is spent. Pin cheap paid coder models instead (deepseek/deepseek-v3.2, qwen/qwen3-coder) billed against org OpenRouter credits, so the OpenRouter slots always have live capacity. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
This was referenced Jul 24, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
PR #619 pinned free tool-capable OpenRouter models, but a live probe (temporary diag workflow on branch openrouter-latency-diag, removed after use) showed every
:freemodel now returnsRate limit exceeded: free-models-per-day-high-balance— the account's shared free-models-per-day quota is exhausted. Worse, OpenCode retries the 429 internally, so each free candidate hangs for its full 5400s timeout: run 30094198328 burned 90min onnvidia/nemotron-3-ultra-550b-a55b:free+ 90min onpoolside/laguna-m.1:freeand still endedexhausted.What
.github/workflows/opencode-review.yml: OpenRouter provider models +OPENCODE_MODEL_CANDIDATESnow pin paid coder models:deepseek/deepseek-v3.2(USD 0.269/0.4 per 1M, ctx 163840, out 65536) andqwen/qwen3-coder(USD 0.3/1, ctx 262144, out 65536). Both are tool-capable; billed against the org's OpenRouter credits (~USD 0.1 per review). Comment block updated.tests/test_opencode_agent_contract.py: candidate list assertions updated (4 sites).scripts/ci/test_strix_quick_gate.sh: 3 candidate assertions updated.Validation
pytest tests/test_opencode_agent_contract.py::test_workflow_provisions_sandbox_tool_and_reviewer_agent tests/test_opencode_agent_contract.py::test_opencode_model_pool_sets_high_effort_for_capable_candidates— 2 passed.toolsand the pinned limits matchtop_providercaps.Developer experience: pool no longer wastes 3h on quota-dead free candidates; OpenRouter slots always have live capacity.
User experience: org PR reviews complete instead of timing out, unblocking the merge queue.