ollama/server
Daniel Hiltgen d26a58557d
MLX: wire up scheduler selected context size for ps (#16918)
In the PS output, expose the scheduler selected size (clamped by model context size) instead of always reporting the model max context.  This will help provide a hint to clients to keep the context size below this value to avoid paging and poor performance on smaller VRAM systems.
2026-06-26 08:47:03 -07:00
..
internal
auth.go
auth_test.go
cloud_proxy.go
cloud_proxy_test.go
create.go
create_test.go
download.go
fixblobs.go
fixblobs_test.go
gemma4_test.go
images.go
images_test.go
inference_request_log.go
logprob.go
model.go
model_caches.go
model_list_cache.go
model_list_cache_test.go
model_recommendations.go
model_recommendations_test.go
model_resolver.go
model_resolver_test.go
model_show_cache.go
model_show_cache_test.go
prompt.go
prompt_test.go
quantization.go
quantization_stub_test.go
quantization_test.go
renderer_resolution.go
routes.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
sched.go MLX: wire up scheduler selected context size for ps (#16918) 2026-06-26 08:47:03 -07:00
sched_test.go
sparse_common.go
sparse_windows.go
test_home_test.go
upload.go