2 results
Show a hover hint with a model's full configuration (input types, context window, max tokens, protocol, base URL, credential env, …) over each model in the DeepSeek Harness web model picker, read from ~/.dsh/settings.yaml.
Patient auto-retry for DeepSeek Harness: honor the upstream gateway's retry_after_seconds on a 429 capacity cooldown instead of giving up after two fast retries, wait out the mislabeled modality 400 a saturated gateway replays from its client-error circuit, and journal every wait as the built-in's own llm/retry records so the console renders its live countdown card. Optional floating countdown badge included.