* Move to backend sampling for MTP draft path
Run top_k(10) on the draft backend. D2H transfers happen only for the top 10 logits
Make backend sampling more robust and fallback to CPU on failure cases, such as with "-sm tensor" or when a backend doesn't support TOP_K.
* Allow sampler chains to be partially offloaded to backend
* Add --spec-draft-backend-sampling argument. Enabled by default.
* save-load-state : refactor into separate phase functions
- Split monolithic main() into 4 self-contained phase functions, each
managing its own context/sampler/batch lifecycle
- Each function tokenizes internally using its local ctx instance
- main() is now a clean orchestrator: init -> run phases -> assert results
- Proper resource cleanup on every exit path (return {} on error)
Assisted-by: llama.cpp:local pi
* save-load-state : use params.out_file instead of separate state_file
- Remove state_file parameter from all phase functions
- Each function accesses params.out_file directly
- Initialize params.out_file in main alongside params.prompt
Assisted-by: llama.cpp:local pi
* save-load-state : use smart pointers for ctx and smpl
- Replace raw llama_context* with llama_context_ptr
- Replace raw llama_sampler* with llama_sampler_ptr
- Remove all manual llama_free() and llama_sampler_free() calls
- Keep llama_batch as raw (managed manually with llama_batch_free)
Assisted-by: llama.cpp:local pi
* save-load-state : add local llama_batch_ptr RAII wrapper
- Add llama_batch_ptr struct holding llama_batch by value
- Calls llama_batch_free() in destructor
- Eliminates all manual llama_batch_free() calls
Assisted-by: llama.cpp:local pi
* save-load-state : replace printf/fprintf with logging macros
- Add log.h include
- Replace fprintf(stderr, ...) errors with LOG_ERR
- Replace fprintf(stderr, ...) info with LOG_TRC
- Replace printf output with LOG
Assisted-by: llama.cpp:local pi
* save-load-state : refactor tests to check results inline
Each follow-up phase now accepts an expected result and performs
the comparison internally instead of collecting results in main().
Assisted-by: llama.cpp:local pi
* save-load-state : improve test output readability
Add phase labels, remove redundant run prefixes, and show
PASS after each test.
Assisted-by: llama.cpp:local pi
* pi : add rule about git signing
* save-load-state : simplify llama_batch_ptr
Change get() to return a reference and remove operator*().
Use batch.get() throughout for consistency.
Assisted-by: llama.cpp:local pi
* save-load-state : extract generate_tokens helper
Factor out the repeated token generation loop into a shared
helper function used by all phases.
Assisted-by: llama.cpp:local pi
* save-load-state : update comments to use test terminology
Replace "Phase" with "Test" and list each test's steps
as bullet points.
Assisted-by: llama.cpp:local pi
* save-load-state : rename test functions
Rename to test_baseline, test_state_load, test_seq_cp_host,
test_seq_cp_device. Update comments and logs accordingly.
Assisted-by: llama.cpp:local pi
* pi : add rule to never git push without confirmation
Assisted-by: llama.cpp:local pi
* common : add model_only option to common_init_from_params
Add bool model_only parameter to skip context creation,
sampler init, and context-dependent setup.
Use in save-load-state to initialize only the model,
with each test creating its own context.
Assisted-by: llama.cpp:local pi
---------
Co-authored-by: ggerganov <ggerganov@users.noreply.github.com>
* spec: support MTP
* fix batch size
* rename files
* cont : simplify (#7)
* MTP: clean-up (#9)
* MTP: clean-up
* review: use llama_context_type instead of llama_graph_type
* review: remove llama_model_has_mtp
* review: fix convert issues
* convert: fix pycheck
* review: formatting
* use `mtp-` for identifying mtp models
* convert: fix mtp conversion
* mtp -> draft-mtp
* remove unused llama_arch
* add need_embd in speculative
* llama: allow partial seq_rm for GDN models for speculative decoding
Currently speculative checkpoint needs to restart from a checkpoint
after some draft tokens are not accepted, this leads to some wastage in
running the target again. This PR adds the ability to rollback upto
`draft_max` by storing the GDN intermediates.
* fix pending state
* vulkan: add GDN partial rollback
* meta: extend check to axis 1
* metal: add GDN partial rollback
Extend the gated delta net kernel to store intermediate states for
partial rollback support on the Metal backend.
- Add K (snapshot slot count) as a function constant
- Read input state from slot 0 of the 3D state tensor
- Write intermediate states to different slots during token loop
- For K=1, maintain backward-compatible single-slot behavior
Ref: 8c05923630
Assisted-by: llama.cpp:local pi
* delta_net_base: use ggml_pad instead of new_tensor
* review: add need_rs_seq
* review: rename part_bounded to n_rs
* review: deslop comments
* review: rename, add asserts
* server : adjust checkpoint logic (#11)
* server : adjust checkpoint logic
* cont : rm asserts
* server-context: fix early exit
* spec : fix compatibility with n-gram and add TODOs (#13)
* metal : cleanup
* llama : fix faulty bitwise check in recurrent memory
* server : disable RS-based MTP in combination with other spec types
* spec : add TODOs
* cont : fix comment
* cont : update comment
* common : fix logic for ngram + mtp compat
* llama-memory: enable checkpointing with partial rollback
* cont: add test-case for loading into a dirty ctx
* llama-memory-recurrent: clear rs_idx in clear
* download: fix mtp path
* llama-arch: fix enorm op
* docs: update docs
* conversion: fix type annotations
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* spec : refactor
* spec : drop support for incompatible vocabs
* spec : update common_speculative_init()
* cont : pass seq_id
* cont : dedup ctx_seq_rm_type
* server : sketch the ctx_dft decode loop
* server : draft prompt cache and checkpoints
* server : improve ctx names
* server, spec : transition to unified spec context
* cont : sync main and drft contexts
* cont : async drft eval when possible
* cont : handle non-ckpt models
* cont : pass correct n_past for drafting
* cont : process images throught the draft context
* spec : handle draft running out of context
* server : fix mtmd draft processing
* server : fix URL for draft model
* server : add comment
* server : clean-up + dry
* speculative-simple : update
* spec : fix n_past type
* server : fix slot ctx_drft ptr
* tools : update readme
* naming : improve consistency
* spec : refactor for multi-sequence speculative context
* cont : prepare params
* cont : prepare params
* spec : support parallel drafts
* server : support parallel drafting
* llama : reuse device buffers when possible
* server, spec : clean-up
* cont : clean-up
* cont : minor
* spec : reset `drafting` flag at the end
* spec : introduce `common_speculative_process()`
* spec : allow for multiple spec types (chain of speculators)
* replace old type field of type common_speculative_type in the
common_params_speculative struct with a vector to allow multiple
types to be specified
* introduce common_get_enabled_speculative_impls(const std::vector<enum common_speculative_type>)
to figure out which implementations the user has enabled
* introduce common_speculative_type_from_names(const std::vector<std::string> & names)
to parse the already user provided spec types
* all speculators run sequentially, best one wins (we verify its drafted tokens)
* maximize expected accepted tokens for current round by calculating the
product between the probability of accepting current token (n_acc_tokens / n_gen_drafts)
and the draft's length
---------
Co-authored-by: Petros Sideris <petros.sideris@nokia.com>
This change refactors the reasoning_budget_message parameter from the
common params into the sampling parameters specifically. It also removes
the reasoning_budget common parameter and standardizes on the existing
reasoning_budget_tokens parameter in the sampling configuration.
Issue: https://github.com/ggml-org/llama.cpp/issues/20429
Original PR: https://github.com/ggml-org/llama.cpp/pull/20297
* tests: allow exporting graph ops from HF file without downloading weights
* use unique_ptr for llama_context in HF metadata case
* fix missing non-required tensors falling back to type f32
* use unique pointers where possible
* use no_alloc instead of fixing f32 fallback
* fix missing space
* tests: allow loading test-backend-ops tests from json
* add error threshold based on op
* add error when file cannot be read
* add graph operator json extraction tool
* add nb parameter for non-contiguous input tensors
* fix view check
* only use view if non-contiguous/permuted, use C++ random instead of rand()
* replace internal API calls with public llama_graph_reserve call
* reduce test description length
* fix nb[0] not getting set for view
* add name to tests
* fix inplace error
* use text file instead of json
* move llama_graph_reserve function to new llama-ext header, move export-graph-ops to tests/
* fix missing declaration
* use pragma once
* fix indent
* fix Windows build
* tests: add end-to-end tests per model architecture
* fixup for rebase
* fix use-after-free in llama-model-loader.cpp
* fix CI
* fix WebGPU
* fix CI
* disable CI for macOS-latest-cmake-arm64
* use expert_weights_scale only if != 0.0f
* comments
* server : support multiple model aliases via comma-separated --alias
* server : update --alias description and regenerate docs
* server : multiple model aliases and tags
- address review feedback from ngxson
- --alias accepts comma-separated values (std::set, no duplicates)
- --tags for informational metadata (not used for routing)
- aliases resolve transparently in router via get_meta/has_model
- /v1/models exposes aliases and tags fields
* regenerate docs
* nits
* server : use first alias as model_name for backward compat
address review feedback from ngxson
* server : add single-model test for aliases and tags
* llama : remove write/read of output ids/logits/embeddings
This commit removes the write/read of output ids, logits and
embeddings from the llama context state.
Refs: https://github.com/ggml-org/llama.cpp/pull/18862#issuecomment-3756330941
* completion : add replying of session state
This commit updates the session handing in the completion tool to handle
the that logits are no longer stored in the session file. Instead, we
need to replay the last token to get the logits for sampling.
* common : add common_prompt_batch_decode function
This commit adds a new function which is responsible for decoding prompt
and optionally handle the saving for session data.
* update save-state.cpp to use llama_state_load_file
This commit updates the save-load-state example to utilize the new
llama_state_load_file function for loading the model state from a file.
And it also replays the last token after loading since this state is now
stored before the last token is processed.
* examples : set n_seq_max = 2 for ctx3
This commit updates the save-load-state example to set the n_seq_max
parameter to 2 when initializing the ctx3 context.
The motivation for this change is that using 1 as n_parallel/n_seq_max
the context only supports one sequence, but the test laster tries to
use a second sequence which results in the following error:
```console
main : loaded state with 4 tokens
main : seq 0 copied, 225760 bytes
main : kv cache cleared
find_slot: seq_id=1 >= n_seq_max=1 Try using a bigger --parallel value
state_read_meta: failed to find available cells in kv cache
```
This seems to only happen for recurrent/hybrid models.
This commit removes two unused functions `common_lcp` and `common_lcs`.
The last usage of these functions was removed in
Commit 33eff40240 ("server : vision support
via libmtmd") and are no longer used anywhere in the codebase.
* initial commit for branch
* simplify constants
* add params to `struct common_params_sampling`, add reference to PR
* explicitly clamp `min_target` and `max_target` to `[0.0, 1.0]`
* add args, rename `queue_size` -> `window_size`
* improved comments
* minor
* remove old unused code from algorithm
* minor
* add power law case to `common_sampler_init`, add sampler name mappings
* clarify behaviour when `window_size = 0`
* add missing enums
* remove `target_range` param, make `target == 1` no-op, cleanup code
* oops, straggler
* add missing parameters in `server-task.cpp`
* copy from author
ref:
https://gist.github.com/MrJackSpade/9be99c7efbba7b95a41377e123b7b069
* remove old debug log, style nit
* fix compiler warning, add commented-out logging per token
* re-write + change parameters + simplify
* oops forgot args.cpp
* fix leftover `window_size`
* add missing values to `common_params_sampling::print()`
* with logging
* does this fix it?
* no, but does this?
* update default decay
* optimize
* fix bad merge
my git skills are lacking
* silence `missing initializer for member`
* update default decay to 0.9
* fix logging
* format (double)
* add power law to the new `samplers` vector
* log sampler init values
* improve logging messages in llama_sampler_power_law
* remove extraneous logging
* simplify target computation
last commit with debug logging!
* remove debug logging, explicitly clamp params at init
* add `use_power_law` flag + logic, minor cleanup
* update `power-law` -> `adaptive-p`
* fix cold start EMA
- `ctx->weighted_sum` is now initialized and reset to `target / (1.0f -
clamped_decay)`
- `ctx->total_weight` is now initialized and reset to `1.0f / (1.0f -
clamped_decay)`
this fixes a "cold start" problem with the moving average
* update `SHARPNESS` constant to `10.0f`
* minor style fixes
no functional changes
* minor style fixes cont.
* update `llama_sampler_adaptive_p_i` for backend sampling (ref: #17004)
* separate into `apply` + `accept` functions
* `pending_token_idx`: switch from `llama_token` to `int32`
functionally identical (`llama.h` has `typedef int32_t llama_token;`),
but its more correct now
* don't transform logits <= -1e9f
* fix masking in backend top-p, min-p
* address review comments
* typo in comments `RND` -> `RNG`
* add docs
* add recommended values in completion docs
* address PR feedback
* remove trailing whitespace (for CI `editorconfig`)
* add to adaptive-p to `common_sampler_types_from_chars`
* server : add arg for disabling prompt caching
Disabling prompt caching is useful for clients who are restricted to
sending only OpenAI-compat requests and want deterministic
responses.
* address review comments
* address review comments
This commit adds the --kv-unified flag to the batched example. This flag
is currently specified in the README.md as required, but is currently
not available as a command line option for the batched example.
The motivation for this is that specifying this flag as the README
instructs, will lead to an error about the flag not being recognized,
and without this option the example fail with the following error:
```console
split_equal: sequential split is not supported when there are coupled
sequences in the input batch (you may need to use the -kvu flag)
decode: failed to find a memory slot for batch of size 4
main: llama_decode() failed
```
* Adding --direct-io flag for model loading
* Fixing read_raw() calls
* Fixing Windows read_raw_at
* Changing type off_t to size_t for windows and Renaming functions
* disable direct io when mmap is explicitly enabled
* Use read_raw_unsafe when upload_backend is available, not functional on some devices with Vulkan and SYCL
* Fallback to std::fread in case O_DIRECT fails due to bad address
* Windows: remove const keywords and unused functions
* Update src/llama-mmap.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---------
Co-authored-by: jtischbein <jtischbein@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* examples : add debug utility/example
This commit introduces a new example named llama-debug which is a
utility that is intended to be used to assist with developing/debugging
a converted model.
The motivation for this utilitiy is to assist in model conversion work
to verify that the model produces the expected outputs. It is intended
to replace logits.cpp in examples/model-conversion.
Example usage:
```console
./build/bin/llama-debug \
-m models/Qwen2.5-0.5B-Instruct.gguf \
--prompt "Hello, my name is" \
--save-logits
...
Model add_bos: false
Input prompt: "Hello, my name is"
Token ids (5):
Hello(9707) ,(11) my(847) name(829) is(374)
Data saved to data/llamacpp-Qwen2.5-0.5B-Instruct.bin
Data saved to data/llamacpp-Qwen2.5-0.5B-Instruct.txt
Prompt saved to data/llamacpp-Qwen2.5-0.5B-Instruct-prompt.txt
Tokens saved to data/llamacpp-Qwen2.5-0.5B-Instruct-tokens.bin
```
For more details about the options available for this example, please
refer to examples/debug/README.md.
* throw runtime error instead of logging error
* remove params.warmup and enable the warmup/nowarmup option
* model-conversion : remove logits.cpp
This commit removes logits.cpp in favor of using llama-debug for
generating logits and embeddings.
* examples : remove model-conversion directory
This was missed in the previous commit.
* model-conversion : add support for saving prompt and token ids
This commit add support for storing the prompt and the token ids for the
prompt when running the original models.
The motivation for this is that this will allow us to compare the prompt
and the tokens generated for the prompt when verifing the converted
model. Currently it is possible that even if the same prompt is used
that the tokens generated are different if there is a difference in the
tokenization between the original and converted model which would
currently go unnoticed (the verification will most likely fail but it
might not be obvious why).
* squash! model-conversion : add support for saving prompt and token ids
fix pyright errors.
* model-conversion : add compare_tokens utility
This commit adds a script to compare token outputs between original and
converted models.
Example usage:
```console
(venv) $ ./scripts/utils/compare_tokens.py pytorch-gemma-3-270m-it llamacpp-gemma-3-270m-it-bf16
Comparing tokens between:
Original : pytorch-gemma-3-270m-it (6 tokens)
Converted: llamacpp-gemma-3-270m-it-bf16 (6 tokens)
✅ All 6 tokens match!
```
And there is a verbose flag that will also print out the prompts:
```console
(venv) $ ./scripts/utils/compare_tokens.py pytorch-gemma-3-270m-it llamacpp-gemma-3-270m-it-bf16 -v
Original model prompt (pytorch-gemma-3-270m-it):
prompt: Hello, my name is
n_tokens: 6
token ids: 2, 9259, 236764, 1041, 1463, 563
Converted model prompt (llamacpp-gemma-3-270m-it-bf16):
prompt: Hello, my name is
n_tokens: 6
token ids: 2, 9259, 236764, 1041, 1463, 563
Comparing tokens between:
Original : pytorch-gemma-3-270m-it (6 tokens)
Converted: llamacpp-gemma-3-270m-it-bf16 (6 tokens)
✅ All 6 tokens match!
```
* model-conversion : add token comparison to verifiction scripts
This commit add the calling of the compare_tokens function in
compare-logits.py and semantic_check.py to ensure that the token ids
that the tokenizers procoduce are the same before proceeding with
verifying the logits/embeddings.
Placing them in the existing scripts instead calling them separately
ensures that the token comparison is always done prior to the
logit/embedding verifications.
Follow up commit/pr could refactor the causal logits verification into
a single script instead of the two that exist now. This would reduce the
code and make it consistent with the embeddings verficiation which only
has a single script.
* debug : use llama_model_n_embd_out
This commit updates the debug example to use the new function
llama_model_n_embd_out instead of llama_model_n_embd.
The motivation for this change is to support late interation retriever
models, like LFM2-ColBert-350M, where the output embeddings are down
projected to a lower dimension.
* debug : add print_usage function
This commit adds a print_usage function that is passed to the
common_params_parse.
The motivation for this is that this enables a specific usage message
which will be printed after all the options, for example:
```console
example usage:
Print tensors:
./build/bin/llama-debug -m model.gguf -p "Hello my name is" --verbose
The tensors to be printed can be filtered with --tensor-filter option.
Save logits/embeddings:
./build/bin/llama-debug -m model.gguf -p "Hello my name is" --save-logits
Add --embedding to save embeddings
```
