diff options
| author | Adam Malczewski <[email protected]> | 2026-06-21 22:07:37 +0900 |
|---|---|---|
| committer | Adam Malczewski <[email protected]> | 2026-06-21 22:07:37 +0900 |
| commit | 4a71e0672f7433f2e29df16fa2865f2702a64db2 (patch) | |
| tree | 92073e809cd18801eb77639829cbf3b5dabc2dbd /packages/kernel/src/runtime | |
| parent | ba9149a6474d0b4cb01c325acea1c83e04f22379 (diff) | |
| download | dispatch-4a71e0672f7433f2e29df16fa2865f2702a64db2.tar.gz dispatch-4a71e0672f7433f2e29df16fa2865f2702a64db2.zip | |
fix(openai-stream): add stream_options.include_usage to all requests
Without stream_options.include_usage, OpenAI-compatible providers omit
the usage field from the SSE stream entirely. Umans returned 0 tokens
for everything; OpenCode's proxy happened to include usage without it.
Now both providers return proper prompt_tokens + completion_tokens.
Note: Umans does not report cache_read_tokens or
prompt_tokens_details.cached_tokens — cache hit rate will be 0% for
Umans regardless. This is a provider limitation, not a parsing issue.
Diffstat (limited to 'packages/kernel/src/runtime')
0 files changed, 0 insertions, 0 deletions