This revamps how we discover GPUs in the system by leveraging the Ollama
runner. This should eliminate inconsistency between our GPU discovery and the
runners capabilities at runtime, particularly for cases where we try to filter
out unsupported GPUs. Now the runner does that implicitly based on the actual
device list. In some cases free VRAM reporting can be unreliable which can
leaad to scheduling mistakes, so this also includes a patch to leverage more
reliable VRAM reporting libraries if available.
Automatic workarounds have been removed as only one GPU leveraged this, which
is now documented. This GPU will soon fall off the support matrix with the next
ROCm bump.
Additional cleanup of the scheduler and discovery packages can be done in the
future once we have switched on the new memory management code, and removed
support for the llama runner.
The context must always be able to store the current batch, so
if the user requests a small context then we should also shrink
the batch to match. This also fixes the TestLongInputContext
test on the new engine. (The old engine already has this behavior.)
* perf: build graph for next batch in parallel to keep GPU busy
This refactors the main run loop of the ollama runner to perform the main GPU
intensive tasks (Compute+Floats) in a go routine so we can prepare the next
batch in parallel to reduce the amount of time the GPU stalls waiting for the
next batch of work.
* tests: tune integration tests for ollama engine
This tunes the integration tests to focus more on models supported
by the new engine.
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.
* 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>
FromFloatSlice and FromIntSlice return an error if the shape doesn't
match the passed data or if memory can't be allocated. Since these
are inputs, the memory being allocated is system memory rather than VRAM.
In many cases, the caller can't really handle the error and panics.
Empty and Zeros directly panic if they can't allocate memory.
This makes things consistent by panicing for the first two cases,
removing a fair amount of error handling code. This is also consistent
with how Go typically handles these situations.
This provides granular information about the backend memory allocations
required by the runner:
- Per backend
- Per layer
- Weights, cache and graph
- Allocation status
This can be used for debugging and validating memory estimates.
Currently, when the backend is created, the tensors are loaded at the
same time, which is a slow operation. This separates them to be two
steps:
- Create backend, including enumerating tensors and memory allocation
- Loading tensor data
This allows more flexibility in managing model loading.
We currently preallocate compute graph memory for the worst case
batch of text tokens. This adds support for doing the same for
images.
Note that image models are more complicated than text models in
how they process their inputs so there may be cases where this
approach isn't completely generic for all models. It covers all
currently supported models though.
For some multimodal models (such as gemma3), we create a single
graph that generates the image embedding and then use this in the
text model. The embedding tensor is completely opaque to the runner.
However, this doesn't work if we need to use the embedding in multiple
batches. This can arise if the embedding is larger than the batch size.
In these cases (as with llama4), we would like to create views that
are more appropriately sized. However, if we do this then the original
source tensor is used in multiple graphs, which isn't allowed. To
avoid that problem, models with this pattern compute the embedding
tensor on first use and recreate the individual views. There is no
longer a single vision and text graph.
This codifies the pattern of separating vision and text graphs. The
logic of computing tensors on demand is moved to the runner, so models
no longer have to worry about this. It also gives the runner visibility
into the multimodal tensors, which is important for memory management.
When we restore a sequence from the cache, we split the prompt into
the already used tokens (stored in the cache) and new tokens that
need to be processed. Currently, the references to the used tokens
are coming from the stored previous sequence.
However, even though we know that the used tokens are semantically
equivalent to the prefix of the prompt, tokens can contain pointers
which are no longer valid. As a result, it is better to get the
used tokens from the prompt, which has currently valid pointers.
This doesn't currently have any impact because it isn't possible
to reuse the pointers (which are tensors) anyways. However, it
becomes an issue once we can.
The correct constant to remove all entries to the end of the sequence
for the Ollama engine is math.MaxInt32. -1 is used by the old engine.
The impact of this is currently minimal because it would only occur
in situations that are not supported by the implemented models or
rarely used options.
Some options listed in api/types.go are not supported in
newer models, or have been deprecated in the past. This is
the first of a series of PRs to clean up the API options
Worst case graph preallocation was disabled by a27462b
"ollamarunner: Temporarily disable worst case graph preallocation"
since it caused crashes with large batches when not using the GPU.
This backports upstream llama.cpp commit f057808
"ggml: Don't assert fail when tensor data changes (#13222)", which
fixes the underlying bug and allows reverting the previous workaround.
The context (and therefore associated input tensors) was not being
properly closed when images were being processed. We were trying to
close them but in reality we were closing over an empty list, preventing
anything from actually being freed.
Fixes#10434
When we later have a large batch running purely on a CPU, this
results the error:
GGML_ASSERT(talloc->buffer_id >= 0)
Disabling this means that we will incrementally reallocate memory
as the graph grows.
Fixes#10410
Currently, the KV cache and graph are lazily allocated as needed.
The cache is fully allocated on first use of the corresponding
layer whereas the graph grows with the size of the context.
This can be an issue if another application allocates more VRAM
after we do our calculations - Ollama will crash in the middle of
inference. If we instead allocate the maximum needed memory at
startup of the runner, we will either succeed or fail at that point
rather than at some surprising time in the future.
Currently, this only generates a worst case batch for text, which
means that vision models may get a partial allocation and continue
to lazily allocate the rest.
No functional change. Many different done reasons can be set at the runner
level, so rather than obsuring them we should return them to the server
process and let it choose what to do with the done reason. This separates
the API concerns from the runner.
The sliding window cache trims entries that are outside the window for
the latest token. This works when we are extending the cache, such as
when the conversation continues. However, if we have a partial overlap
in conversation (including the BOS tokens), then we resume from a past
point in the conversation and the needed tokens are no longer stored
in memory. This verifies that the new window overlaps with the old one
before reusing the cache.
Co-authored-by: Jesse Gross <jesse@ollama.com>
When truncating inputs to the the context window at the beginning of
a sequence, we remove the minimum amount possible. However, this
may cause us to truncate to the middle of a set of inputs that
the model specified should not be split up. To avoid this, we
need to remove the rest of the partial batch.
Clear KV cache when shift operation is not supported by model.
Added KvCacheCanShift() check to handle models that can't perform cache shifts,
falling back to full cache clear while preserving logical token history to
maintain expected behavior when context window fills up.
If we have an error after creating a new sequence but before
finding a slot for it, we return without releasing the semaphore.
This reduces our parallel sequences and eventually leads to deadlock.
In practice this should never happen because once we have acquired
the semaphore, we should always be able to find a slot. However, the
code is clearly not correct.
With the llama runner, we can generate up to NUM_PARALLEL batches
at once, which will then get broken up to into individual batches
to get executed by llama.cpp (i.e. we add up to 2048 tokens and
this gets split into 4 batches of 512 tokens at default settings).
This splitting can improve parallelism on multi-GPU systems because
the individual batches can move though the pipeline without blocking
on the first one to fully complete. However, we don't yet support
this in the Ollama runner, partially because it makes it hard to
enforce model-specified batch constraints, which didn't exist
previously.
The result is that we will try to execute the full, unsplit batch.
This could result in out of memory or insufficient KV cache space
errors.
This triggers batch breaking when the total inputs from all sequences
exceeds the batch size, rather than per-sequence. In order to ensure
fairness, it also reintroduces round-robinning around sequences so
that we don't let one busy sequence starve the others.
Currently the runner computes the kv size needed and creates a
cache of that size. This is the context size times number of
parallel sequences.
Cache implementations can make better decisions about their memory
usage, so instead pass in the required capacity, number of sequences
and maximum batch size. For now, the causal cache just uses this to
compute the size in the same way as before.
This enables the runner to report progress back to the Ollama server,
both for showing status to the user and also to prevent the server
from killing the runner if it thinks things have stalled.
Most of the infrastructure was already there, this extends it to
be available to the backends.
Rather than directly giving the input data to models, we can
pass a tensor instead. In the short term, this saves some duplicated
code.
Longer term, we will want to overlap setting up the next batch with
processing of the current one. In this case, we will only have the
shape of tensor but it will not be loaded with data at the time of
graph generation. By passing only a tensor to models now, we set up
this possibility and prevent them from relying on data that they won't
have in the future.
Although the same could be done for Positions and Outputs, in some
cases we either need the raw input data or don't use them at all.
Therefore, for now we leave them as they are and allow models to
convert them to tensors as needed.
Models can specify that a group of inputs need to be handled a single
batch. However, context shifting didn't respect this and could trigger
a break anyways. In this case, we should instead trigger a context
shift earlier so that it occurs before the grouped batch.
Note that there still some corner cases:
- A long prompt that exceeds the context window can get truncated
in the middle of an image. With the current models, this will
result in the model not recognizing the image at all, which is
pretty much the expected result with truncation.
- The context window is set less than the minimum batch size. The
only solution to this is to refuse to load the model with these
settings. However, this can never occur with current models and
default settings.
Since users are unlikely to run into these scenarios, fixing them is
left as a follow up.
We do not need to bypass the prompt caching in the ollama runner yet, as
only embedding models needed to bypass the prompt caching. When embedding
models are implemented they can skip initializing this cache completely.
Currently there is a single context per sequence, shared all by
all multimodal inputs. Since we build a vision encoder graph per
image, with a large number of inputs we can eventually hit the
maximum number of graph nodes per context.
This changes to use a separate context for each image, ensuring
that available resource limits are consistent.
Models may require that a set of inputs all be processed as part
of the same batch. For example, if an image has multiple patches
with fully connected attention between them, we should not split
the batch in the middle of an image.
Fixes#9697
