fix: prevent cloud reasoning models from exhausting default max_tokens

- Send max_tokens = 32768 for non-local providers so reasoning models have room for both chain-of-thought and answer.

- Add per-provider thinking_budget option for endpoints (currently SiliconFlow) that cap reasoning tokens separately.

- Accept bare JSON arrays from /v1/models; some OpenAI-compatible endpoints omit the {"data": [...]} envelope.

1 files changed, 14 insertions, 0 deletions

diff --git a/CHANGELOG.md b/CHANGELOG.md
index d3df977..6e9a831 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -34,6 +34,20 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   `model`. Token counts come from the stream's final usage chunk and power the
   cost readout; the model column records which model actually replied, so a
   switched conversation prices each turn correctly.
+- Per-provider `thinking_budget` option, for endpoints (currently SiliconFlow)
+  that cap chain-of-thought tokens separately from the final answer. Unset
+  providers skip the key entirely so other endpoints do not receive an unknown
+  parameter.
+
+### Fixed
+
+- Model listing crashed when a provider's `/v1/models` endpoint returned a
+  bare JSON array instead of the standard `{"data": [...]}` envelope. The
+  parser now accepts both shapes.
+- Cloud chat requests no longer rely on the provider's default `max_tokens`,
+  which could be too small for a reasoning model. Non-local providers now send
+  `max_tokens = 32768`, leaving room for both chain-of-thought reasoning and
+  the answer.
 
 Tool calling and reasoning output vary between providers. Web search on a
 cloud model may not work as reliably as it does with a local router.