Celes Renata eead4f1381 fix: disable thinking mode on vLLM path with chat_template_kwargs ...

The thesis rewriter uses vLLM (not Ollama) in production. The previous
fix only added think=False to the Ollama payload. For vLLM's
OpenAI-compatible API with Qwen3 models, thinking mode is disabled via
chat_template_kwargs: {enable_thinking: false} in the request body.

2026-04-29 16:04:04 +00:00

..

adapters

fix: position sync now reconciles — removes positions broker no longer holds

2026-04-29 12:02:57 +00:00

aggregation

feat: signal math upgrade — probabilistic, regime-aware scoring pipeline

2026-04-29 11:41:48 +00:00

api

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

extractor

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

ingestion

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

lake_publisher

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

parser

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

recommendation

fix: disable thinking mode on vLLM path with chat_template_kwargs

2026-04-29 16:04:04 +00:00

risk

feat: auto-clamp buy orders to fit within position limits instead of hard-rejecting

2026-04-28 14:20:44 +00:00

scheduler

fix: pipeline stop now halts all workers and flushes queues

2026-04-29 07:59:35 +00:00

shared

feat: signal math upgrade — probabilistic, regime-aware scoring pipeline

2026-04-29 11:41:48 +00:00

symbol_registry

fix: resolve 6 integration test failures

2026-04-20 04:30:13 +00:00

trading

fix: return open_position_count in trading status for dashboard display

2026-04-28 19:27:40 +00:00