Jesse Gross 32568531bd create: read draft architecture from its config.json ...

Previously the draft architecture was hardcoded to
Gemma4AssistantForCausalLM. Read it from the draft model's config so
any draft architecture can be packaged.

2026-05-22 09:32:09 -07:00

..

client

create: read draft architecture from its config.json

2026-05-22 09:32:09 -07:00

create_test.go

mlx: Gemma4 MTP speculative decoding (#15980)

2026-05-05 08:55:04 -07:00

create.go

mlx: Gemma4 MTP speculative decoding (#15980)

2026-05-05 08:55:04 -07:00

dtype.go

mlx: Support NVIDIA TensorRT Model Optimizer import (#15566)

2026-04-27 18:28:10 -07:00

gemma4_test.go

Keep Gemma4 router projection in source precision (#15613)

2026-04-15 15:04:23 -07:00

gemma4.go

Keep Gemma4 router projection in source precision (#15613)

2026-04-15 15:04:23 -07:00

imagegen.go

create: Clean up experimental paths, fix create from existing safetensor model (#14679)

2026-04-07 08:12:57 -07:00

laguna_test.go

New models (#15861)

2026-04-28 11:50:12 -07:00

laguna.go

New models (#15861)

2026-04-28 11:50:12 -07:00

qwen35.go

create: Clean up experimental paths, fix create from existing safetensor model (#14679)

2026-04-07 08:12:57 -07:00