This commit modifies the graph building logic to immediately continue
when a token has already been sampled by the backend sampler.
It also updates the test for backend temporary sampling to include
top-k and distribution samplers in the chain to verify that they are not
producing any logits (they are not run).
This commit fixes the implementation of the temperature-based sampler
for the case when the temperature is set to zero. This now correctly
selects the most probable token by masking out all other tokens in the
logits.
In addition to match the algorithm proposed in the original
[paper](https://arxiv.org/abs/1904.09751), this resolves the edge-case
where `max_p is > top_p` for a single logit, where the mask would
otherwise be empty (and we thus sample from the whole vocabulary with
equal likelihood)
This commit changes the logging level for output buffer reallocations
in the llama_context::output_reserve function from INFO to DEBUG.
The motivation for this is that it currently logs to info and when
enabling verbose logging for llama-cli this will get mixed with the
output, for example:
```console
What is the capital of Sweden?output_reserve: reallocating output buffer from size 0.58 MiB to 1.74 MiB
1. Stockholm
2\. Helsinki
Based are the options
1. Stockholm
Explanation: Stockholm is the capital of
...
```
* ggml : add GGML_SCHED_NO_REALLOC option to disable reallocations in ggml_backend_sched
Enabled in ggml-ci for testing.
* llama : update worst-case graph for unified cache
* ci : disable op offload in some tests
* fix spelling
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
As we only support static graphs for the time and we don't know the size
of the output of top-p, we have to do value-scaling same as for min-p
operator.
Further improvements can be applied to the unit-test (i.e. check for
equivalence of top_p happening on backend with top_p happening on cpu)
and also by constructing candidates and sorting those as opposed to
reversing the sort of the logits (this would be arange +
get_rows instead of argsort + get_rows)
Fix llama-save-load-state which currently fails by handling the case
when batch.logits is nullptr (like when loading state) by allocating
space for all outputs as CPU logits.
* Qwen3 Next - cleaned up version
* Whitespaces and stuff
* Correct minor errors
* Update src/llama-model.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Misc. fixes.
* Clean up code, add missing hybrid qualifier
* Did someone transpose the SOLVE_TRI result matrix? Perhaps...
* Whitespace
* Proper tensors for cb calls
* Use llama-graph.h vertical alignment
* BROKEN: chunking
* Set new tensors as inputs.
* Proper chunk logic
* It's the circle of life...
* More shenanigans for n_seq > 1
* Nail in the coffin?
* Fix Windows build
* Eh, one fails on Windows, the other fails on Mac... just use general capture.
* quant : cleanup
* model : cleanup
* qwen3 : cleanup
* cont : cleanup
* cont : cleanup
* ggml : revert change
* qwen3 : cleanup
* cont : cleanup
* Readd cmath
* qwen3 : fix typo
* Update convert_hf_to_gguf.py
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Usual suspects
* fix my bad suggestion
---------
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit updates the backend sampling implementation to support
intermixed usage of backend and CPU samplers within the same batch.
The initial implementation was developed as an all-or-nothing solution:
either perform backend sampling for the entire batch, or perform CPU
sampling for the entire batch.
The motivation for this change is to support batches with mixed
sequences. For example, we may have a backend sampler configured for
sequence 0, while sequence 1 in the same batch uses CPU sampling. This
was not supported in the initial implementation.
This issue manifested in llama-server with the webui: decoding with
backend samplers would work initially, but after changing to CPU
sampling, a slot (sequence) could still be using a backend sampler.
This meant that logits in output_reserve would not be allocated,
resulting in an error.
The solution in this commit inspects the batch to determine which
sampling modes are needed and allocates buffers accordingly. However,
there is a known inefficiency: when we have intermixed backend/CPU
samplers in the same batch, we currently copy all logits to the host,
even for sequences using backend samplers.
Added test_backend_cpu_mixed_batch to verify correct behavior with
mixed backend/CPU samplers in a single batch, including dynamic
sampler switching between decode calls.
This commit adds a check to skip the output reordering logic when
n_outputs == 1. With a single output token, the data is trivially
sorted and the reordering code is currently doing unnecessary work
(resetting and rebuilding output_ids to the same values).
The motivation for this change is improved code clarity and avoiding
confusion when debugging. While the performance impact is probably
negligible, this unnecessary work happens on every decode call in
llama-server when processing batches with single-token outputs.
This commit adds a debug log statement in the llama_sampler_sample
to indicate when a backend sampler has selected a token for a given
index.
The modification helps in tracing the sampling process and understanding
the flow of control when backend samplers are used.
This commit fixes the issue where both sampled tokens and logits/probs
were not being copied correctly from the backend to the host when
multiple backend samplers were used.
A test for this scenario has also been added to ensure that both types
of data are copied correctly when different backend samplers are
employed.
* Detect GigaChat3-10-A1.8B as deepseek lite
Hardcodes checking number of layers to detect if lite version of deepseek.
* Add commnent identifying deepseek lite variants
deepseek lite variants include DeepSeek-V2-Lite, GigaChat3-10B-A1.8B
This commit introduces a sampling_info struct to encapsulate all
backend sampling related data within the llama_context class.
It also updates to use more descriptive names for sampled tokens and
candidates in the backend sampler ggml data structure.
* Fix DoS / integer overflow
* Remove optional, use INT64_MAX instead as placeholder value (it's technically -1, so it fits :)
* White space
* Actually, since it's unsigned, use UINT64_MAX
This commit updates common/sampler.cpp set_logits and
src/llama-sampling.cpp llama_sampler_sample to always populate the
logits field when backend sampled probabilities are available.
The motivation for this is that this ensure that CPU sampler always have
access to the logits values even when probabilites have been produced by
backend samplers.
Test 'Q4_K_M' quantization on https://huggingface.co/pfnet/plamo-2-translate
The 'suffix_to_score' size is 193510, it needs 19 memory allocation with final
capacity 262144 to hold the value, if not preserve the memory.
Signed-off-by: Haiyue Wang <haiyuewa@163.com>
This commit precomputes and caches the full-vocab token id list in
llama_context's constructor, so llama_get_backend_sampled_token_ids_ith
always returns a valid pointer.
The motivation for this is that this enables both common/sampling.cpp
and src/llama-sampling.cpp can simplify their logic.
Not all backends samplers that process logits need to set the
sampled_tokens_id as they may not change the order of the logits, for
example the temperature sampler only scales the logits but does not
change their order. Simliar the logit bias sampler only adds bias to
specific token ids but does not change the order of the logits. In
these cases there will not be a device to host copy of the sampled
token ids, and this is the use case where having this precomputed
list is useful.
This commit adds support for performing sampling operations on the
backend (e.g. GPU) as part of the model computation graph.
The motivation for this feature is to enable sampling to be performed
directly on the backend as part of the computation graph being executed,
allowing for some or all of the sampling to be done on the backend.
For example, the backend sampler chain might select/sample a token
directly in which case only the sampled token needs to be transferred
from device memory to host memory.
It is also possible for the backend samplers to perform filtering of
the logits, or compute and filter the probability distribution, in
which case only the filtered logits or probabilites need to be
transferred back to system memory for further processing by CPU
samplers.
Currently the backend sampling works in a similar manner to how
pooling works, it is a function that is called by build_graph and the
sampler operations become part of the models computation graph.
* fix : Dangling pointer for non-empty trigger words in llama_sampler_init_grammar_impl (#17047)
* Replace 'static' workaround, with keeping variable in scope for longer
* Create std::array directly and pass into llama_grammar_init_impl
* Add back the trigger pattern
* Missed array include
* CUDA: add fused rope
* move k forward_expand up
* create helper function instead of re-using params
* make assert statement more in line with comment
* rope_norm: coalesced writes to global mem
When compiling llama.cpp in Yocto, it fails QA checks because the generated so files aren't versioned. This applies a version to all generated so files, allowing the package to build without errors.
* feat(memory): Only fail partial erasure of recurrent tail
The recurrent state is always assumed to be the state as of the last update
from the final token in the sequence. When doing a partial erasure, if the
range does not include the final token, the erasure can be considered a
success since any memory used for the sequence prior to the final token
(which is no memory) has been successfully removed.
There is one potential case that this doesn't address which is the pruning
of cache to remove sensitive data from the context. This wouldn't work for
attention cache partial removal (in the middle) either since the KV state
is linearly-dependent and states in later sequence positions would still be
based on the state from the sensitive data, even if that data is no longer
cached, so I don't think this is relevant, but it is worth noting that the
semantics of this change for a partial erasure in the middle of the cache
are essentially "my context is already compressed" and not "all trace of
the removed tokens has been removed."
https://github.com/ggml-org/llama.cpp/issues/16768
Branch: HybridContextShift-16768
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* fix(main): Check the output of seq_rm for prefix matching
This prefix matching is explicitly attempting to remove the tokens at the
end of the sequence that don't match. This is the operation that can't be
performed on a recurrent cache due to the state being updated in place, so
if this removal fails, we need to clear the whole cache.
https://github.com/ggml-org/llama.cpp/issues/16768
Branch: HybridContextShift-16768
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* fix(memory): Fix condition for partial erasure failure if p0 > pos
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
Co-authored-by: compilade <git@compilade.net>
* style: Fix extra parens
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* fix(main.cpp): Set n_matching_session_tokens to 0 on cache clear
https://github.com/ggml-org/llama.cpp/issues/16768
Branch: HybridContextShift-16768
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
---------
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
Co-authored-by: compilade <git@compilade.net>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* kv-cache : pad the size of the small SWA cache for performance
* context : pad the total context to 256
* cont : future-proof the swa pad
* server : adjust test params to new logic
* server : support unified context across slots
* cont : fix speculative decoding initialization
* context : fix n_ctx_per_seq computation
* server : purge slots one by one
* tests : add unified cache server tests
* llama : update per-seq context computation
* test-thread-safety : handle tiny training context of the input model
* server : fix server_tokens clear()
* server : use 4 slots + unified KV by default
* llama : add note about context size queries
* cont : update todos [no ci]
* context : do not cap the size of the context
* tests : adjust parameters to be CI friendlier
* context : add warning
* Added GGUF mappings for CogVLM model
* Add tensor mapping for CogVLM visual encoder
* Add CogVLM to conversion script, no vision part yet
* Added CogVLM vision model to conversion script
* Add graph for CogVLM CLIP model
* Add graph for CogVLM
* Fixes for CogVLM. Now compiles.
* Model now runs
* Fixes for cogvlm graph
* Account for graph context change after rebase
* Changes for whitespace
* Changes in convert script according to comments
* Switch CogVLM LLM graph to merged QKV tensor
* Use rope_type variable instead of direct definition
* Change CogVLM CLIP encoder to use SWIGLU
* Switch CogVLM CLIP to use merged QKV
* Apply rebase edits and remove ggml_cont call that is now unnecessary
* clean up
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Georgi Gerganov
Daniel Bevenius
Herman Semenoff
Adrien Gallouët
Piotr Wilkin (ilintar)
Gilad S.
Oliver Simons
Xuan-Son Nguyen
Aman Gupta
Diego Devesa
Aaron Teo
william pan
ubergarm
Haiyue Wang
Bartowski
Marek Hradil jr.
o7si
Mike Abbott
Sigbjørn Skjæret
Gabe Goodhart
Johannes Gäßler
Li Pengzhan
Giuseppe Scrivano
JJJYmmm
Tianyue-Zhao