Commit graph

10 commits

Author SHA1 Message Date
Daniel Hiltgen
cecac8bf1b
mlxrunner: add Inkling 2 Mini (#19799)
* mlxrunner: add Inkling 2 Mini

* mlx: fuse Inkling 2 Mini residual mixing on Metal

Fuse RMS normalization and query scoring in one kernel. Apply the
softmax weights in a second kernel to avoid storing per-query products.

* create: expose Inkling 2 Mini context length in model metadata

Copy model_max_length to max_position_embeddings during import so
ollama show reports the checkpoint's context length. Reject configs
that specify conflicting limits.

* mlxrunner: reduce Metal dispatches in Inkling 2 Mini decode

Reuse the prefill mask shader to compute relative position bias in one
dispatch for single-query attention. On M3 Ultra with Metal 3, compute
the loop controller's gain and forget projections in one matrix multiply.
2026-11-20 18:38:44 -08:00
Daniel Hiltgen
0b243a141a
mlxrunner: choose default parallelism from available memory (#19785)
When OLLAMA_NUM_PARALLEL is unset, select 1, 2, 4 or 8 slots
using available memory and estimated per-client storage at the
model's context limit.

Preserve explicit settings and retain the serial default under
memory pressure, for unknown cache layouts and on other backends.
2026-11-20 07:13:42 -08:00
Daniel Hiltgen
73ee71ddcc
mlxrunner: adapt saved history to memory pressure (#19678)
* mlxrunner: adapt saved history to memory pressure

Retain Metal prefix history using observed memory headroom and execution
workspace rather than a fixed 8 GiB allowance. Release unused buffers and
cold history when system compression grows, preserving reusable paths and
active requests.

Cover pressure reclamation, snapshot ownership, memory observation, and
cacheless request isolation.

* address comments

* address comments
2026-11-19 15:41:42 -08:00
Daniel Hiltgen
b8db8d5b5c
mlxrunner: reuse saved prefixes across request slots (#19677)
Reuse completed prefix checkpoints from another slot, including one still
serving a request. Restore into recipient-owned state and copy full-attention
history directly without materializing an intermediate donor snapshot.

Cover donor lifetime, pending captures, cancellation, media and drafting
boundaries, merged history, and cached evaluation.
2026-11-16 20:54:35 -08:00
Daniel Hiltgen
ab9295d03c
mlx: support concurrent generation (#19197)
* mlx: batch requests through a shared scheduler

Use one request queue and scheduler for generation, decisions, embeddings
and speech, including parallelism one. Share compatible model operations
while keeping request caches, sampling state and response streams independent.

Interleave prompt chunks with decoding and retain speculative decoding when
generation runs alone. Cover isolation, cancellation, slow readers, cache
reuse and mixed requests with unit and integration tests. Remove the stale
llama-server architecture restriction on parallel requests.

* mlx: prevent cache snapshots from retaining batch peers

Copy each request’s initial batched K/V row so a sliding-cache snapshot cannot retain the whole batch allocation while accounting for only one row.

Exclude grammar compilation waits from prompt timing and reuse row-selected attention masks across layers.

Return errors for requests rejected before admission and strengthen numerical and compiled-speech regression checks.

* address comments
2026-11-10 13:24:12 -08:00
Daniel Hiltgen
66ab8482e0
MLX, MLX-C: version bump - prompt speed up for gemma/flint/laguna (#19291)
* MLX, MLX-C: version bump

The MLX-C pin now provides its own bindings for gather_qmm global_scale
and fast::gated_delta_update, so those two carried patches drop out.
Carry 0001-mlx-c-sdpa-window-size instead: sliding-window attention
added a window_size argument to fast::scaled_dot_product_attention
(ml-explore/mlx#4734) that MLX-C has not bound yet, to be regenerated
from MLX-C once the upstream binding PR merges. The SDPA wrappers gain
a window size argument; all call sites default to the full context.

* mlxrunner: dispatch sliding-window prefills to the native SDPA window

Sliding layers no longer materialize a causal∩sliding mask for prefill
chunks past the window: resolve() dispatches {causal, window} natively
when the rotating cache's L>1 concat has linearized the ring and the
mask is otherwise pure causal, via a new MaskApplier.NativeWindow hint.
Vision relaxations, padding, model arrays, and decode keep the
materialized path. No model changes; sinks compose natively.
2026-10-25 21:25:58 -07:00
Daniel Hiltgen
f580b5484a
mlx: add Flint model support (#19181)
Preserve source tool JSON alongside decoded API values so Flint can retain
numeric types and schema order in prompts. Existing API marshaling is unchanged.

Assemble ModelOpt expert banks during safetensors import, and reject direct
Flint NVFP4 creation before conversion. Other models retain their existing
import and quantization paths.

Add optional learned sinks and device causal masks to shared MLX attention.
Existing callers retain the no-sink and host-mask paths; Flint opts into the
new behavior.
2026-10-21 07:45:50 -07:00
Daniel Hiltgen
a5b659e61b
mlx: compact restored recurrent state after speculative rollback (#18955)
* mlx: compact restored recurrent state after speculative rollback

Per-token recurrent captures can leave a restored state pointing into a larger buffer. If a request ends before another model step, the live cache retains that buffer even though it accounts for only the selected state.

Track when a restored state may share that backing buffer. At request close, compact marked states in one batch before retaining them in the prefix cache. Clear the marker when a normal model step replaces the state, and cover restoration and buffer detachment in tests.

* address comments
2026-10-12 09:28:22 -07:00
Daniel Hiltgen
a00d083bb5
mlx: System one support (#18701)
* mlxrunner: restore System One scoring

Carry the MLX implementation and its tests from the original System One
commit. The API/compiler and llama-server implementation already exist on
the release branch, so retain those versions. Adapt the optional model test
to the renamed decision package and its explicit rendering/context setup.

(cherry picked from commit 26e546ddb3)

* Enable MLX System One and add regression coverage

* mlxrunner: add Laya support for System One

Import Laya checkpoints and implement bidirectional decision scoring.

* mlxrunner: release cached buffers after System One scoring

* decision: match Tev1 and Nimble publisher prompt formats

Select Tev1 prompts using renderer metadata and fix JSON formatting, with regression coverage for GGUF and MLX.

* mlx: support Clef and Clef Flash decision models

Add model import, quantization safeguards, and text/image scoring with the Clef joint decision head.

* support vision capabilities for decision models

* Make Clef scoring cancellable between prefill chunks

Preserve joint-head inputs and test cancellation, cleanup, and recovery.

* mlx: separate decision preparation and readout from forward

Let serial and batched scoring share model-owned preparation and readout
without duplicating architecture-specific logic in the runner.

* mlx: add Strands Decider support

* implement caching

---------

Co-authored-by: ParthSareen <parth.sareen@ollama.com>
2026-10-03 13:17:21 -07:00
Jesse Gross
2e036e7cdf mlx, mlxrunner: move the MLX engine out of x/
The MLX runner is the only Go inference runner left and is no longer
experimental, so its packages leave x/. The bindings become a top-level
mlx package beside the carried patches in mlx/compat, mirroring how
llama/ holds the llama.cpp integration, and the runner becomes mlxrunner
with the architectures nested under the package they implement.
Subpackages move with their parent unless listed.

  x/mlxrunner/mlx            mlx
  x/internal/mlxthread       mlx/mlxthread
  x/internal/mlxthreadtest   mlx/mlxthread/mlxthreadtest
  x/internal/mlxtest         mlx/mlxtest
  x/quant                    mlx/quant
  mlx/compat/*.patch         mlx/compat/mlx-c   (MLX patches go in mlx/compat/mlx)
  x/mlxrunner                mlxrunner
  x/models/nn                mlxrunner/nn
  x/models/<arch>            mlxrunner/model/<arch>
  x/mlxrunner/imports.go     mlxrunner/model/architectures   (new package)
  x/create                   create
  x/safetensors              fs/safetensors
  x/tokenizer                mlxrunner/tokenizer

Every package keeps its name, so the Go changes are the import path
rewrites the moves force, and the CMake, Dockerfile, CI cache keys, drift
check and Darwin payload script follow the new paths. Four edits are not
paths: the runner's blank architecture imports become the package
mlxrunner/model/architectures, so the list to extend for a new model sits
beside the architecture directories; a depguard rule keeps the two test
harnesses out of non-test code, as the x/internal placement used to; the
CI change filter's two entries for the long-deleted x/imagegen/mlx now
name the bindings' CMake project and the carried patches, so a change to
either builds the payload; and the tokenizer parity test reads its
fixtures from its own testdata instead of walking out of x/.

x/server and x/imagegen/manifest stay for the next two commits.
2026-09-16 14:06:08 -07:00