b10658
spec : add DFlash2 support (local convolution + candidate selector) (#27342) (#27816) spec : add DFlash2 support (local convolution + candidate selector) (#27342) support DFlash2 Add p_min in DFlash2
Knowledge catalogue
spec : add DFlash2 support (local convolution + candidate selector) (#27342) (#27816) spec : add DFlash2 support (local convolution + candidate selector) (#27342) support DFlash2 Add p_min in DFlash2
model: add Qwen3.8-Flash-Next (qwen4exp) (#27742) gguf: add qwen4exp (Qwen3.8-Flash-Next) arch and converter Adds the GGUF-side plumbing for HF model_type qwen4_exp: MODEL_ARCH.QWEN4EXP plus tensors f
What's Changed ollama launch claude no longer shows a warning about Claude connectors by @hoyyeva in #18878 ollama launch codex no longer warns about service tiers by @hoyyeva in #18879 Paused backgro
model : support for Prism Bonsai 2 27B (#29600) Runtime support for Prism Bonsai 2 27B Assisted-by: Claude Code address prism hadamard runtime feedback move hadamard tensors into method, define folded
model : add LiquidAI/d1-omni-600M decision model (#30114) model : add LiquidAI/d1-omni-600M decision model Assisted-by: Claude Opus 5.5 mtmd : keep conformer GLU sigmoid on CUDA Assisted-by: Claude Op
model : add LiquidAI/d1-3B decision model (#30110) model : add LiquidAI/d1-3b decision model mtmd : read LFM2 image resize algo from GGUF Assisted-by: Claude Opus 5.5 common : rename decision type d1
llama : add a GPU cache for MoE experts kept in host memory (#29887) llama : add a GPU cache for MoE experts kept in host memory Assisted-by: Claude use llama_moe_cache_ptr Website: https://llama.app
vulkan : fix TOP_K for +inf/NaN inputs and k = 1 on negative values (#30107) The bucket search in topk_nary_search.comp started from the range [0, 0xFF800000), which ends just below the ordered-uint m
metal: fix NaN in mul_mm_id when activations exceed f16 range (#26223) test-backend-ops: reproduce MUL_MAT_ID NaN for activations beyond f16 The Metal mul_mm_id path narrows src1 to half for the simdg
Ollama v0.23.0 was released on May 3, 2026 , with v0.23.1 appearing as a patch release shortly after. The v0.23.0 release added support for Claude Desktop through Ollama Launch, with Claude Cowork and
ggml-openvino: update to 2026.4.1, optimize performance, expand ops, improve device listing. (#29852) ggml-openvino : Qwen3.5 MoE perf (#312) Squash of ravi9#312: ggml-openvino: add detailed inference
server : fix laya abort by limiting n_batch to n_ubatch (#29903) server : fix laya abort by limiting n_batch to n_ubatch Fixes #29902 Assisted-by: Claude fix(review) : rm tests, embeddings cond Websit
ggml-quants : avoid invalid rounding in qkx3 scale search (#29817) ggml-quants : avoid invalid rounding in qkx3 scale search The imatrix scale search can produce an infinite, NaN, or otherwise out-of-
chat : honor json_schema in Ling 3.0 parser (#29813) chat : honor json_schema in Ling 3.0 parser Ling 3.0 only built a grammar for tool calls and did not handle inputs.json_schema, so response_format
add GLM-5.3-Flash (GLM5-Next) support (#27773) Rebase GLM-Next support onto master, and migrate to llama-memory-hybrid-idx Add initial MTP support Merge branch optimizations. Reduce allocated compute
model : support classifier_pooling for rerankers (#29627) model : support classifier_pooling for ModernBERT rerankers Assisted-by: Claude Opus 5.5 model : read classifier pooling type in load_hparams
ggml : add BF16 unary, GLU, binary and scale ops (CPU, CUDA) (#29675) ggml : add BF16 unary, GLU, binary and scale ops (CPU, CUDA) ggml-cpu : use per-op _bf16 functions for BF16 unary and GLU ops Assi
qwen4exp: fix seq_cp, block position keying, mtmd input, cuda abort, add tests (#27941) qwen4exp: follow up fixes -kvu NaN collapse fix Assisted-by: Claude indexer cache ext.x/ext.y restore fix Assist
vulkan : dequant q8_0 KV once in coopmat1 (#25494) vulkan : dequant q8_0 KV once in coopmat1 Assisted-by: Claude (Opus 4.8) vulkan : fall back instead of aborting when FA scratch exceeds maxStorageBuf
model : GraniteSWAForCausalLM / GraniteMoeSWAForCausalLM (#25505) feat(convert): Add conversion for GraniteSWAForCausalLM Branch: GraniteSWAForCausalLM AI-usage: full (Bob, OpenCode + Qwen3.6-35b) Sig
feat: add GLM5Next MTP, optimize (#29928) llama : add GLM5-Next NextN (MTP) graph Build the GLM5-Next multi-token-prediction head as graph_mtp: the NextN block embeds enorm(tok)+hnorm(h) through eh_pr
model: add Kimi-K3 text model (#26185) model: add Kimi-K3 text model Hybrid KDA (linear) + MLA (full) attention as in Kimi-Linear-48B, plus five things that architecture does not have: cross-layer res
What's Changed launch: add Muse Code integration by @dhiltgen in #17594 model/renderers: match Muse Glimmer reasoning template by @dhiltgen in #17732 launch: add DeepSeek Harness integration by @Parth
chat : pass reasoning_effort to template chat: add reasoning_effort to common_chat_templates_inputs Store OpenAI Chat Completions reasoning_effort and make it available to jinja templates (with model
Muse Glimmer Muse Glimmer is now available on all platforms. Muse Glimmer can power coding agent applications such as Claude Code, Codex, Pi and more, as well as long-running personal assistants such
SYCL: add oneMKL GEMM flash attention for XMX-accelerated prompt proc… (#25025) SYCL: add oneMKL GEMM flash attention for XMX-accelerated prompt processing fattn-mkl: fix interleaved dst layout in nor
model : use exact GELU for ModernBERT encoders (#30108) model : use exact GELU for ModernBERT encoders Assisted-by: Codex model : keep tanh GELU aliases on ggml_geglu Assisted-by: Claude Opus 5.5 mode
metal : gate mul_mm_id src1 rescale behind ggml_prec (#29029) metal : gate mul_mm_id src1 rescale behind ggml_prec Assisted-by: Claude Fable 5.1 ggml-webgpu: reject MUL_MAT_ID when src1 precision is F
OpenVINO: optimize stateful decode and GPU MoE inference (#28638) exclude GPU/NPU failing POOL_2D case Fix pool case ggml-openvino: fix stateful decode for Gemma-4 per-layer-type head sizes ggml-openv
rpc : hash-cache only weights (#28789) rpc : hash-cache only weights ggml_backend_rpc_buffer_set_tensor and ggml_backend_rpc_set_tensor_async hashed every transfer above HASH_THRESHOLD and let rpc-ser
CUDA/HIP: Flash Attention tuning (gfx1201) (#28102) HIP: enable mma FA for head size 256 on RDNA4, tune configs Assisted-by: Claude Assisted-by: Codex HIP: prefer whole-tile FA grids over stream-k on
What's Changed mlx: fix mac assumptions on linux/windows by @dhiltgen in #17898 mlx update by @dhiltgen in #17886 lint fixes by @dhiltgen in #17897 app: add claude desktop app by @ParthSareen in #1789
vulkan: sparse flash attention for quantized K/V (#29639) vulkan: sparse flash attention for quantized K/V Assisted-by: Claude vulkan: single-scan sparse FA index compaction The compaction ran one wor
vulkan: fix stale prealloc_y reuse across flash attention and soft_max (#29591) Assisted-by: Claude Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52808438
vulkan: disable large matmul tile on Samsung GPUs with 32KB shared memory (#28531) Assisted-by: Claude Opus Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/
llama-mmap : avoid a second full-size copy of each tensor with direct-io (#29749) Assisted-by: Claude Co-authored-by: Pranesh Gonegandla pgonegandla@nvidia.com Website: https://llama.app Attestations:
common : add LLM-jp-4.1 Harmony dialect handler (#29681) LLM-jp-4.1 uses the GPT-OSS format, but its tokenizer decodes a space after every special token and parallel tool calls are separated by <|end|
chat : add message delimiters to the DeepSeek V3.2/V4 parser (#29008) chat : add message delimiters to the DeepSeek V3.2/V4 parser Assisted-by: Claude Co-authored-by: Sigbjørn Skjæret sigbjorn.skjaere
gguf : align the data section relative to the GGUF start, not the file (#28993) gguf : align the data section relative to the GGUF start, not the file gguf_init_from_file_ptr reads a GGUF from the cur
vulkan : add command-buffer debug labels for GPU profilers (#28101) vulkan : add command-buffer debug labels for GPU profilers Co-authored-by: gabby-zy z2262718160@gmail.com Assisted-by: Claude Code v
vulkan: handle larger batch sizes (>4) efficiently for IQ3_S mat-vec (#27449) vulkan: handle larger batch sizes (>4) efficiently for IQ3_S mat-vec when NUM_COLS > 4. 5x perf at n=8 Assisted-by: Claude
kv-cache : optimize restoring non-contiguous cells (#27991) kv cache : batch state restore scatter reads per contiguous run When restoring state into non-contiguous destination cells (e.g. a prompt-ca
grammar : parse - in char classes as literal hyphen (#27591) grammar : accept '-' escape in character classes gbnf_escape_char_class() escapes '-' as '-' but parse_char() rejected that escape, so gene
chat : fix LFM2 tool call arg name prefix ambiguity (#26960) Assisted-by: Claude Opus 5 Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled)
Muse Glimmer Note: Muse Glimmer is currently available via initial support via Ollama's MLX engine on Apple Silicon. Additional support and optimizations for Apple Silicon, NVIDIA, AMD, and other plat
Extended SYCL oneDNN SDPA to non-FP16 KV caches (Q4_0–Q8_0 and FP32) (#25874) sycl: extend oneDNN SDPA to Q4_0-Q8_0 and F32 KV caches Extends the oneDNN SDPA path (PR #25222) to handle non-F16 KV cach
model: add NextN/MTP speculative decoding support for GLM_DSA (GLM-5.2) (#25980) model: add NextN/MTP speculative decoding support for GLM_DSA (GLM-5.2) Adds GLM-5.2 NextN/MTP as a --spec-type draft-m
What's Changed launch: keep Claude Code channels available by @hoyyeva in #17210 cmd: remove dead agent prompt wrappers by @ParthSareen in #17227 agent: reorder working directory instruction by @Parth
v0.24.0-rc1 is a pre-release version focusing on improvements to Ollama server caching and the desktop launch experience, including plan-aware model gating and disabling Claude Desktop launch. The rel
vulkan: int8 coopmat1 matmul implementation for AMD RDNA3 and RDNA4 (#27952) vulkan: add int8 coopmat quantized matmul shader apply scales inline use scalar sums probe and directly access coopmat valu
CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#28552) Recreated from #24546 Co-authored-by: Carl Philipp Klemm carl@uvos.xyz CUDA: pick MMQ tile size against ncols_opt set on t
chat : split specialized parsers into common/parsers (#27764) chat : split specialized parsers into common/parsers Move the 14 dedicated template parsers out of chat.cpp into one file each under commo
llama : re-reserve the sched when the nextn extraction flags change (#30020) The speculative MTP init enables NextN extraction on the target and draft contexts after both were created and their schedu
server : remove the built-in UI's service worker when the UI is not served (#29565) With --path or --no-ui, /sw.js returned 404, and a 404 does not remove a service worker, so browsers kept showing th
batch: migrate speculative, mtmd and server to batch_ext (#29385) adapt common add common_batch wip wip: spec cont common_speculative_process server_batch to use common_batch rm some stale calls Assis
CUDA: fuse RMS_NORM + SCALE into one kernel (#29393) #28068 builds the GDN q/k l2norm as ggml_scale(ggml_rms_norm(x, eps/n), 1/sqrt(n)). This adds 2 SCALE nodes per GDN layer, 96 extra kernel launches
vulkan: hide internal symbols to prevent duplicate-dlopen state destruction (#29139) Since #28732 our internal symbols are exported. A duplicate copy dlopened and dlclosed by ggml_backend_load_all() t
vulkan: add IQ4_XS MMQ/MMV matmul kernels (#28415) vulkan: optimize IQ4_XS matmul kernels Assisted-by: OpenAI Codex vulkan: address IQ4_XS review nits drop the dead LOAD_VEC_A != 8 branch in the IQ4_X
cuda : fix CUB argsort corruption caused by in-place keys (#28389) argsort_f32_i32_cuda_cub called the one-shot DeviceRadixSort::SortPairs API with d_keys_in == d_keys_out (temp_keys, temp_keys). CUB'
vulkan: raise the hoisted row-id limit for mul_mat_id from 256 to 1024 experts (#28501) vulkan: raise the hoisted row-id limit for mul_mat_id to 512 experts The expert-count shader (count_experts.comp