digital-garbageandClaude Sonnet 5.5 4549ffc2b3 feat: Diffusion Model Loader says when fp32 compute is picked over bf16 weights ...

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

2026-10-17 03:19:36 +03:00

..

test_attention_path.py

loaders: three the attacker found, and the silence around one of them

2026-08-29 03:41:04 +03:00

test_compute_dtype.py

loaders: compute_dtype cast the input and left the weights behind

2026-08-29 12:06:45 +03:00

test_gguf_support.py

fix: modules no longer import each other (shared taste helper is a plain file); UI gates match real ids and are tested; boot cannot stall; empty pipelines refused; truncated .gguf listing says so (attacker round 15)

2026-10-15 22:10:20 +03:00

test_int8_convrot.py

feat: int8_convrot option on the diffusion model loader (H3 bf16 file -> int8 + rotation at load)

2026-10-04 14:09:24 +03:00

test_loader_behaviour.py

feat: Diffusion Model Loader says when fp32 compute is picked over bf16 weights

2026-10-17 03:19:36 +03:00

test_loader_schemas.py

loaders: the four FunPack loaders, as V3 nodes

2026-08-29 03:32:06 +03:00

test_sla_attention.py

fix: dense steps '0' means none, as the tooltip and the earlier commit say (was flagged as a bad token)

2026-10-10 16:15:51 +03:00

test_weight_report.py

feat: loader logs the file's real weight format (int8 file at bf16 says int8); warns when a cu12 torch on Blackwell halves int8 speed

2026-10-10 18:34:58 +03:00