Skip to content

Tags: ollama/ollama

Tags

v0.34.3-rc1

Toggle v0.34.3-rc1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
server: allow registry cross-host redirects among allowlisted hosts (#…

…18533)

v0.34.3-rc0

Toggle v0.34.3-rc0's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
api: expose model thinking levels and defaults (#18473)

v0.34.2

Toggle v0.34.2's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
x/transfer, server: tighten redirect handling for registry requests (#…

…18512)

Redirects for registry and blob transfers now validate the target scheme
and resolved addresses before following, re-check DNS on each redirect,
and do not follow redirects that switch an https session to plain http.
The --insecure option continues to relax address checks for private
registries but not scheme checks.

v0.34.2-rc3

Toggle v0.34.2-rc3's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
cli: add first-run onboarding shared with the desktop app (#18495)

v0.34.2-rc2

Toggle v0.34.2-rc2's commit message
mlxrunner: Release freed KV buffers during speculative decode

The decode loop releases MLX's pool of freed buffers every 256 generated
tokens, which is also how often the KV cache grows and drops its previous,
smaller buffers. The check fires only when the token count lands exactly on
a multiple of 256. Speculative decoding emits several tokens per round, so
most rounds step over the boundary and the pool is never released. Each
growth at a long context leaves several GB of buffers that no later
allocation can reuse, so the runner's footprint keeps climbing over a long
generation until the system runs out of memory.

We now release the pool whenever a round crosses a multiple of 256 tokens,
which is what a single-token round already did. With qwen3.8:27b-mlx at a
98k-token context on a 128 GB machine, a long speculative generation
previously grew the runner past 90 GB and panicked the kernel; it now stays
flat at 30 GB.

v0.34.2-rc1

Toggle v0.34.2-rc1's commit message
mlxrunner: lay out model by contract, checkpoint and construction

model is one package with three jobs: the contract between the runner
and the architectures, the opened checkpoint, and building nn layers
from checkpoint tensors. Its files did not say which was which. base.go
carried the folded package's name over the interfaces and the registry,
root.go held the safetensors header scan next to Root, and quant.go
mixed the nvfp4 global-scale helpers with quant parameter resolution.

base.go becomes model.go, named for what it holds. root.go keeps Root
and Open; TensorQuantInfo and the header scan join quant.go, so
everything the checkpoint says about quantization is read and resolved
in one file. The global-scale helpers move to globalscale.go with their
tests. Root.Close, a no-op with one caller, goes. No code changes
otherwise.

v0.34.2-rc0

Toggle v0.34.2-rc0's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
llama.cpp: version bump b10969 (#18446)

llama.cpp build changes resulted in duplicate symbols between libllama and libmtmd.  This moves the compat patch into libllama with exported symbols.

v0.34.1

Toggle v0.34.1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
docs: refresh getting started guides (#18450)

v0.34.1-rc2

Toggle v0.34.1-rc2's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
API: Deprecate typical_p (#18448)

Drop support for creating new models with typical_p parameters, while
retaining support for existing GGUF models with the setting.

v0.34.1-rc1

Toggle v0.34.1-rc1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
mlx: add mlx patch to docker build context (#18440)