The thinking parser will automatically transition to being a
pass-through if non-whitespace is seen before an opening tag. However,
we weren't clearing the buffer after the first non-whitespace input, so
in practice the first token would be emitted twice.
Added a test that demonstrated this, and then fixed the bug.
With old memory estimates, it's currently impossible to load more
than one model at a time when no GPUs are available. This is because
the check for whether we need to evict a model looks to see if all
layers of the new model can be loaded onto GPUs, which is never true
if there are no GPUs. Before the memory management changes, there
was a special code path for CPU-only systems.
This problem does not exist with new memory estimates.
Fixes#11974
0x007e is a tilde and was getting adjusted (+0x00a2) to 0x0120 in the
encode, but then in the decode it was getting adjusted down (-0x0100) to
0x0020. The boundary for the +0x00a2 case has been adjusted to fix this
Fixes: #11966
Flash attention kernels require the mask of the KV cache be a F16
rather than an F32. We can use the GGML operation ggml_cast to do
this rather than doing it ourselves, which allows reuse of a
preallocated buffer in the graph rather than allocating a new one
for each batch. This improves token generation performance with
flash attention by 10-30% (with gpt-oss). This also makes performance
with flash attention better than without it, as expected.
We dump out our best memory estimate after we complete processing
for any reason, including errors. This is helpful for finding what
what stopped us in error conditions but in some cases we might not
have gotten even the first result yet.
Fixes#11957
In <https://github.com/ollama/ollama/issues/11704#issuecomment-3177380197>
I noticed that hyphens in function names could possibly cause the model
to become confused. Later in that issue I found other explanations, but
at a minimum tool names with spaces in them are confusing to the model
because of the prompt format.
In this change I create a mapper that converts arbitrary tool names into
valid typescript identifiers. It's a little overly strict in that it
doesn't allow all unicode characters that might be valid in ts
identifiers, but it's still very permissive. Since mappings aren't
reversible, we must temporarily store this mapping in order to unmap it
if the model comes back with a call. We also handle the case where
multiple mappings collide into the same mapping and append a counter to
the end to make them unique
* test: improve scheduler/concurrency stress tests
The scheduler test used to use approximate memory figures and would often
over or under shoot a systems capcity leading to flaky test results.
This should improve the reliability of this scenario by leveraging
ps output to determinie exactly how many models it takes to
trigger thrashing.
The concurrency test is also refined to target num_parallel + 1 and handle
timeouts better.
With these refinements, TestMultiModelConcurrency was redundant
* test: add parallel generate with history
TestGenerateWithHistory will help verify caching and context
are properly handled while making requests
* test: focus embed tests on embedding models
remove non-embedding models from the embedding tests
This changes the memory allocation strategy from upfront estimation to
tracking actual allocations done by the engine and reacting to that. The
goal is avoid issues caused by both under-estimation (crashing) and
over-estimation (low performance due to under-utilized GPUs).
It is currently opt-in and can be enabled for models running on the
Ollama engine by setting OLLAMA_NEW_ESTIMATES=1. Behavior in other
cases is unchanged and will continue to use the existing estimates.
if there's no transformation to the tensor and the input and output
types match, copy directly into the writer. also read from a bufio with
a 32K buffer
* TEMPORARY: Update the llama.cpp upstream to my fork's Granite Four branch
This will be redone once my branch is merged upstream in llama.cpp
* feat: Update all patches
There are a number that are no longer needed at all:
- 0003-embeddings: Embeddings entirely overhauled on master
- 0008-ensure-KV-cache-is-fully-defragmented: KV caching entirely
overhauled on master
- 0019-metal-add-mean-kernel-14267: Merged upstream
- 0020-CUDA-add-mean-operation-14313: Merged upstream
* feat: Sync llama.cpp and ggml
* fix: Update rsync-filter for all moved/new/removed files
* fix: Add files missing from sync
* fix: Update ggml rsync-filter for new ggml-cpu/arch subdirs
* fix: Add ggml files missing from sync
* fix: Narrow llama.cpp rsync-filter to not include mtmd main tool cpp files
* fix: Remove mtmd main cpp files
* fix: Add missing include in sampling_ext.cpp
* fix: Update llama.go to use mtmd instead of clip/llava
* fix: Add patch for mtmd_input_text
* chore: Ignore *.patched in the patch directory
* fix: Fix support for arch-specific ggml-cpu source files with new arrangement
In https://github.com/ggml-org/llama.cpp/pull/13892, all arch-specific
implementations were split out into a nested tree structure under
ggml-cpu/arch. This conflicts with standard CGO layout where all
arch-specific source files are expected to live in the same directory as
the parent go module and use suffixes based on GOOS and GOARCH. As such,
there were really two options for getting this to work:
1. Add a patch on top of the GGML sync to rearrange the files to match the
GO layout convention
2. Use CGO directives to conditionally include the nested source files in
the compilation units
This commit does (2) in order to minimize the set of changes needed on top
of the upstream file layout. To get this to work, there are two key things
needed:
1. In cpu.go, #cgo directives are added to explicitly set __${GOARCH}__ in
the preprocessor directives
2. In arch-impls.c|cpp, use an #ifdef | #elif defined | #endif chain to
explicitly include the .c|.cpp files for the given architecture from the
nested directory
* fix: Use mtmd_helper to correctly load the bitmap for the image
* fix: Apply patch for mtmd_text_input
* fix: Add missing stb to llama.cpp rsync-filter
* fix: Add sync'ed stb vendored header
* fix: Use c++17 and include vendor for go wrapper modules
* fix: Update patch 0015 for upstream implementation of uuid
* feat: Bump to the latest tip of the branch
* fix: Update patches for bump
* feat: Bump back to the cenral repo and point at the latest master
This includes granite 4 and a number of other model architectures!
* fix: Revert changes to ggml export GPU UUID patch
* fix: Add patch for GGML_VERSION and GGML_COMMIT constants
* feat: Sync all patched code
* build: Include cmake/common.cmake in ggml sync
* build: Add top-level include for GNUINstallDirs in CMakeLists.txt
This is used to populate CMAKE_INSTALL_BINDIR
* fix: Add a patch to avoid power throttling API on non-msvc windows builds
* fix: Sync patch changes for ggml-cpu.c
* feat: Bump llama.cpp to 4a4f42
This picks up support for Kimi K2 and PLaMO-2
* feat: Sync llama.cpp
* fix: Handle multi-chunk image encodings from mtmd
* fix: Re-number patches after merge with `main`
* feat: Bump to 41e78c in the makefile
* fix: Fix Solar and argsort/copy patches after bump
* fix: Remove Gemma3n CUDA Graphs patch
It was implemented upstream:
https://github.com/ggml-org/llama.cpp/pull/14741
* feat: Sync llama.cpp / ggml after latest bump
* build: Remove unnecessary CFLAGS definitions in cpu.go
* fix: Remove unnecessary additions in the rsync-filter
* fix: Remove unused vendored code for chat template parsing
* Revert "fix: Remove Gemma3n CUDA Graphs patch"
This reverts commit d724caced3.
* fix: Update 0020 CUDA Graphs for gemma3n to keep both llama.cpp and ollama fixes
https://github.com/ollama/ollama/pull/11195#issuecomment-3137312394
* fix: Sync ggml-cuda.cu after keeping both style cuda graph fixes for gemma3n
* unwind mxfp4 patch
Prepare to bump ggml with their impl for mxfp4
* bump
* fix windows build error
* Convert tensors at load time
Repack the mxfp4 tensors as ggmls kernels expect them to be.
* convert mlp bf16 to f32
* buffer the conversion better
* reshape earlier
* openai swiglu
* add ids
* split qkv, gate_up
* fix nested alt tags
* fast attention
* remove debug messages
* fix lint
* remove redundant test
* remap values only if source/target are different
* add back i32->i32 copy
* refactor cpu quants
* clean up vendor
* update patch instructions
* clean up patches
* remove webgpu
* update mem
* also handle gpt-oss
* revert convert changes
---------
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
Co-authored-by: Gabe Goodhart <ghart@us.ibm.com>
Co-authored-by: Daniel Hiltgen <daniel@ollama.com>
We prefer the nvcuda library, which reports driver versions. When we
dropped cuda v11, we added a safety check for too-old drivers. What
we missed was the cudart fallback discovery logic didn't have driver
version wired up. This fixes cudart discovery to expose the driver
version as well so we no longer reject all GPUs if nvcuda didn't work.
Prior to this change our official binaries contained both JIT PTX code and
the cubin binary code for our chosen compute capabilities. This change
switches to only compile the PTX code and rely on JIT at runtime for
generating the cubin specific to the users GPU. The cubins are cached
on the users system, so they should only see a small lag on the very
first model load for a given Ollama release. This also adds the first
generation of Blackwell GPUs so they aren't reliant on the Hopper PTX.
This change reduces the ggml-cuda.dll from 1.2G to 460M
Some AMD GPUs do not provide UUIDs and report only "XX". In these
cases, we should use the ordinal ID as an alternate identifier.
This is the same as we always need to do on Windows for AMD.
In addition, this prints out the ID for each GPU when enumerating
them for easier debugging in the future.
This patch modifies Ollama to allow grouping GPUs to memory-fit to the requested model, instead of the former algorithm of using one GPU distributing over all available GPUs.
Benefits:
- Lower amount of (PCIe-)bus communication between GPUs - especially when they are not very high speed
- Allowing unallocated GPUs to get into power-saving mode.
- Significantly reduce VRAM allocation when using more than 2 GPUs in a system
- Due to the reduced memory allocation, you can run more models simultaneously.
Callers can set a backend buffer type to be no-alloc, meaning that
it does not allocate memory for tensors or operations. This can
be used for calculating memory requirements. Tensors and graphs
must be recreated with no-alloc set to false before loading data.
Defaults to false for newly created backend buffer types.
For many backend data structures, GGML defines a typedef of a pointer
type and returns these from functions. In most cases, CGo understands
that these are interchangable but some parts of Go (such as generics)
think they are two different types. We should prefer the form that
GGML uses.
gpt-oss works best with a context length of at least 8k. However,
for GPUs with limited amount of VRAM, there is a significant
performance hit to this increased context. In these cases, we
switch to the Ollama default of 4k
We were missing passing along thinking if content was nil (as opposed
to empty string)
Also added a test for content not being passed, which was the real cause
of <https://github.com/ollama/ollama/issues/11704>, since with the way
`Content` is typed, not passing it and empty string are distinct
