Qwen/Qwen3.8-Flash lists $0.16 input, $0.47 output, and $0.016 cache read
per million tokens, and z-ai/glm-5.3-flash lists $0.15, $0.50, and $0.03;
the official pricing page documents no cache-write rate for either model.
Document the incremental reasoning-delta streaming, live display pricing,
and dynamic model fetch in CHANGELOG.md. Add a Pricing section to
README.md explaining the MODEL_COSTS overlay and how to update prices.
Add tests/test-pricing.ts to assert that known models have non-zero
pricing, promotional deals are documented in comments, and Claude models
include cache pricing. Fix test-pi-local to match against combined
stdout+stderr since pi may output model listing to either stream.
Thomas Byr