iomgaa 1f13eb18ab feat: let one call ask for a different tier than its source defaults to ...

The three-layer priority (call > source > enable_thinking sugar > silence)
now lives in one pure function, thinking.effective_effort(). The assembly
guard and the request hot path used to each carry their own inline copy of
the sugar conversion; two copies of the same judgement drift into the worst
shape there is - passes at assembly, raises at runtime.

The guard now also honours effort_fallback, so a source that opted into
nearest is no longer sentenced at assembly for a tier it could have mapped.

2026-09-05 02:15:25 -04:00

..

__init__.py

feat: add in-memory limiter and breaker satisfying backend contracts

2026-07-20 06:54:07 -04:00

test_backpressure.py

docs: correct how a wait-mode call actually dies on a dead source

2026-08-20 01:00:47 -04:00

test_cache.py

fix: close the failure modes review found in the new code

2026-08-26 02:37:24 -04:00

test_client.py

feat: let one call ask for a different tier than its source defaults to

2026-09-05 02:15:25 -04:00

test_config.py

feat: let a source name its reasoning tier, and say so when it contradicts itself

2026-09-05 01:56:24 -04:00

test_embedding.py

feat: carry caller dimensions down the embedding chain

2026-08-17 09:53:37 -04:00

test_errors.py

feat: let every gateway error carry what the gateway said

2026-08-16 06:01:38 -04:00

test_health_selector.py

fix: address M2.5 verifier findings before merge

2026-07-21 21:10:02 -04:00

test_http_error_body.py

feat: add the single summarizer for HTTP error bodies

2026-08-16 06:03:26 -04:00

test_monkey_ocr.py

test: make the 429 red line actually catch its violation

2026-08-16 06:50:50 -04:00

test_ocr_client.py

feat: carry caller dimensions through the OCR chain

2026-08-17 11:34:39 -04:00

test_openai_compat.py

feat: give zhipu, moonshot, anthropic and google a wire of their own

2026-09-05 00:42:58 -04:00

test_package.py

feat: refuse an impossible tier with the cheapest one that model does have

2026-09-05 01:36:30 -04:00

test_ports.py

feat: record the reasoning verdict in telemetry

2026-08-26 00:29:26 -04:00

test_pricing.py

feat: cap telemetry bodies at a configurable length

2026-08-19 13:39:17 -04:00

test_providers.py

feat: give zhipu, moonshot, anthropic and google a wire of their own

2026-09-05 00:42:58 -04:00

test_redis_key_layout.py

fix: reparent governance backend failures under GatewayUnavailableError (issue #7)

2026-08-06 04:53:52 -04:00

test_retention_tool.py

fix: refuse the sandbox rather than quietly running it as the superuser

2026-08-26 11:59:51 -04:00

test_retry.py

feat: carry the reasoning verdict through to LLMResponse

2026-08-26 00:03:28 -04:00

test_soak_corpus.py

fix: make RPM jitter exemption side-aware with one-shot room

2026-07-21 07:42:54 -04:00

test_soak_scoreboard.py

fix: address M3 verifier findings on soak invariants and probe tests

2026-07-22 02:21:33 -04:00

test_sources.py

feat: add gateway client with env-driven assembly

2026-07-20 07:47:05 -04:00

test_streaming.py

feat: add gateway client with env-driven assembly

2026-07-20 07:47:05 -04:00

test_structured.py

fix: make the sampling invariant test actually enforce the constraint

2026-07-31 22:01:51 -04:00

test_telemetry.py

fix: close the failure modes review found in the new code

2026-08-26 02:37:24 -04:00

test_thinking.py

feat: let one call ask for a different tier than its source defaults to

2026-09-05 02:15:25 -04:00

test_types.py

feat: carry the reasoning verdict through to LLMResponse

2026-08-26 00:03:28 -04:00

test_usage_source_domain.py

feat: cap telemetry bodies at a configurable length

2026-08-19 13:39:17 -04:00