jmorganca d0f38a915a llama/compat: add gpt-oss and lfm2 handlers ...

gpt-oss: rename arch "gptoss" -> "gpt-oss" (incl. KV prefix), inject
the missing `expert_feed_forward_length` from the ffn_gate_exps shape,
and rename `attn_out`/`attn_sinks`/`ffn_norm` tensors to upstream's
`attn_output`/`attn_sinks.weight`/`post_attention_norm`. Also remove
the library/gpt-oss -> dhiltgen/gpt-oss redirect now that the compat
shim handles it directly.

lfm2: rename `output_norm.weight` -> `token_embd_norm.weight` and fix
a stale `lfm2.feed_forward_length` (some Ollama blobs claim 12288 on
a model whose ffn_gate is [2048, 8192]) by reading the real value off
the ffn_gate tensor shape.

Adds two helpers to compat-util: `copy_kv` (type-preserving generic
KV copy) and `rename_kv_prefix` (bulk-copy every KV with a given
prefix to a new prefix). Old keys are left in place — harmless because
the loader queries by exact name and only the new prefix matters.

Tested locally: gpt-oss:20b and lfm2.5-thinking now load + generate
coherently against an unmodified upstream llama-server build.

2026-04-20 09:29:34 -07:00

..

internal

…

auth_test.go

…

auth.go

…

cloud_proxy_test.go

…

cloud_proxy.go

…

create_test.go

…

create.go

…

download.go

…

fixblobs_test.go

…

fixblobs.go

…

gemma4_test.go

…

images_test.go

…

images.go

…

inference_request_log.go

…

logprob.go

…

model_resolver_test.go

…

model_resolver.go

…

model.go

…

prompt_test.go

…

prompt.go

…

quantization.go

…

renderer_resolution.go

…

routes_cloud_test.go

…

routes_create_test.go

…

routes_debug_test.go

…

routes_delete_test.go

…

routes_generate_renderer_test.go

…

routes_generate_test.go

…

routes_harmony_streaming_test.go

…

routes_list_test.go

…

routes_options_test.go

…

routes_request_log_test.go

…

routes_test.go

…

routes_web_experimental_test.go

…

routes.go

…

sched_test.go

…

sched.go

…

sparse_common.go

…

sparse_windows.go

…

test_home_test.go

…

upload.go

…