Celes Renata 0437943863 fix: reduce vLLM default max_tokens to 4096, update model to AxionML/Qwen3.5-9B-NVFP4 ...

The model's max_model_len is 16384 — requesting 32768 output tokens
caused HTTP 400 from vLLM. 4096 is a safe default for extraction output.

2026-04-23 19:49:34 +00:00

..

adapters

feat: stage-isolated infrastructure — separate Postgres DBs, Redis DBs, and MinIO bucket prefixes per stage

2026-04-19 22:20:03 +00:00

aggregation

fix: dedup recommendation queue at aggregation level — prevent duplicate ticker+window flooding

2026-04-17 22:11:59 +00:00

api

feat: pipeline on/off toggle with per-stage Helm control

2026-04-21 00:21:53 +00:00

extractor

fix: add extract() method to VLLMClient for extraction pipeline compatibility

2026-04-23 19:32:33 +00:00

ingestion

fix: track last_published_at per source to avoid re-fetching same articles — applies to both news_api and macro_news

2026-04-16 18:12:12 +00:00

lake_publisher

feat: stage-isolated infrastructure — separate Postgres DBs, Redis DBs, and MinIO bucket prefixes per stage

2026-04-19 22:20:03 +00:00

parser

feat: competitive intelligence & historical pattern matching layer

2026-04-14 19:42:48 +00:00

recommendation

fix: skip LLM thesis rewrite for informational/suppressed recs to prevent queue buildup

2026-04-17 19:29:41 +00:00

risk

fix: risk engine blocking sell orders on over-concentrated positions

2026-04-22 02:07:24 +00:00

scheduler

feat: beta deploys all services with pipeline toggle defaulting to OFF

2026-04-21 03:54:00 +00:00

shared

fix: reduce vLLM default max_tokens to 4096, update model to AxionML/Qwen3.5-9B-NVFP4

2026-04-23 19:49:34 +00:00

symbol_registry

fix: resolve 6 integration test failures

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

trading

fix: risk engine blocking sell orders on over-concentrated positions

2026-04-22 02:07:24 +00:00