* llama-server : implement universal assisted decoding
* Erase prompt tail for kv-cache
* set vocab_dft_compatible in common_speculative
* rename ctx_main to ctx_tgt
* move vocab_dft_compatible to spec struct
* clear mem_dft, remove mem
* detokenize id_last for incompatible models
* update comment
* add --spec-replace flag
* accept special tokens when translating between draft/main models
* Escape spec-replace
* clamp draft result to size to params.n_draft
* fix comment
* clean up code
* restore old example
* log common_speculative_are_compatible in speculative example
* fix
* Update common/speculative.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Update common/speculative.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Update common/speculative.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit adds support for the `embd_normalize` parameter in the
server code.
The motivation for this is that currently if the server is started with
a pooling type that is not `none`, then Euclidean/L2 normalization will
be the normalization method used for embeddings. However, this is not
always the desired behavior, and users may want to use other
normalization (or none) and this commit allows that.
Example usage:
```console
curl --request POST \
--url http://localhost:8080/embedding \
--header "Content-Type: application/json" \
--data '{"input": "Hello world today", "embd_normalize": -1}
```
Currently if RPC servers are specified with '--rpc' and there is a local
GPU available (e.g. CUDA), the benchmark will be performed only on the
RPC device(s) but the backend result column will say "CUDA,RPC" which is
incorrect. This patch is adding all local GPU devices and makes
llama-bench consistent with llama-cli.
* Mtmd: add a way to select device for vision encoder
* simplify
* format
* Warn user if manual device selection failed
* initialize backend to nullptr
* imatrix : allow processing multiple chunks per batch
* perplexity : simplify filling the batch
* imatrix : fix segfault when using a single chunk per batch
* imatrix : use GGUF to store imatrix data
* imatrix : fix conversion problems
* imatrix : use FMA and sort tensor names
* py : add requirements for legacy imatrix convert script
* perplexity : revert changes
* py : include imatrix converter requirements in toplevel requirements
* imatrix : avoid using designated initializers in C++
* imatrix : remove unused n_entries
* imatrix : allow loading mis-ordered tensors
Sums and counts tensors no longer need to be consecutive.
* imatrix : more sanity checks when loading multiple imatrix files
* imatrix : use ggml_format_name instead of std::string concatenation
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* quantize : use unused imatrix chunk_size with LLAMA_TRACE
* common : use GGUF for imatrix output by default
* imatrix : two-way conversion between old format and GGUF
* convert : remove imatrix to gguf python script
* imatrix : use the function name in more error messages
* imatrix : don't use FMA explicitly
This should make comparisons between the formats easier
because this matches the behavior of the previous version.
* imatrix : avoid returning from void function save_imatrix
* imatrix : support 3d tensors with MUL_MAT
* quantize : fix dataset name loading from gguf imatrix
* common : move string_remove_suffix from quantize and imatrix
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* imatrix : add warning when legacy format is written
* imatrix : warn when writing partial data, to help guess dataset coverage
Also make the legacy format store partial data
by using neutral values for missing data.
This matches what is done at read-time for the new format,
and so should get the same quality in case the old format is still used.
* imatrix : avoid loading model to convert or combine imatrix
* imatrix : avoid using imatrix.dat in README
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* initial commit for handling extra template kwargs
* enable_thinking and assistant prefill cannot be enabled at the same time
* can set chat_template_kwargs in command line
* added doc
* fixed formatting
* add support for extra context in generic template init
* coding standard: common/chat.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* coding standard: common/chat.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Apply suggestions from code review
coding standard: cosmetic changes
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* fix merge conflict
* chat.cpp: simplify calls to apply to ensure systematic propagation of extra_context (+ the odd existing additional_context)
* normalize environment variable name
* simplify code
* prefill cannot be used with thinking models
* compatibility with the new reasoning-budget parameter
* fix prefill for non thinking models
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: Olivier Chafik <olivier.chafik@gmail.com>
Mistral Small 2506 models using Pixtral vision encoder were running out
of GPU memory when processing images larger than 1024x1024 pixels due to
exponential memory growth from unlimited image size.
This fix applies the same 1024x1024 limit used by Qwen2VL models to
prevent OOM issues while maintaining compatibility with existing models.
Add no_warmup parameter to cmd_params struct and command-line parsing to allow users to skip warmup runs before benchmarking.
- Add no_warmup boolean field to cmd_params struct
- Add --no-warmup command-line argument parsing
- Add help text documentation for the new flag
- Wrap existing warmup logic in conditional check
- Maintain full backward compatibility (warmup enabled by default)
Addresses #14224
* webui: fix sidebar being covered by main content
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
* webui: update index.html.gz
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
---------
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
* llama : deprecate llama_kv_self_ API
ggml-ci
* llama : allow llama_memory_(nullptr)
ggml-ci
* memory : add flag for optional data clear in llama_memory_clear
ggml-ci
* threading: support for GGML_SCHED_PRIO_LOW, update thread info on Windows to avoid throttling
We talked about adding LOW priority for GGML threads in the original threadpool PR.
It might be useful for some cases to avoid contention.
Latest Windows ARM64 releases started parking (offlining) the CPU cores
more aggresively which results in suboptimal performance with n_threads > 4.
To deal with that we now disable Power Throttling for our threads for the NORMAL
and higher priorities.
Co-authored-by: Diego Devesa <slarengh@gmail.com>
* threading: disable SetThreadInfo() calls for older Windows versions
* Update tools/llama-bench/llama-bench.cpp
Co-authored-by: Diego Devesa <slarengh@gmail.com>
---------
Co-authored-by: Diego Devesa <slarengh@gmail.com>
* Replace alert and confirm with custom modals. This is needed as Webview in VS Code doesn't permit alert and confirm for security reasons.
* use Modal Provider to simplify the use of confirm and alert modals.
* Increase the z index of the modal dialogs.
* Update index.html.gz
* also add showPrompt
* rebuild
---------
Co-authored-by: igardev <ivailo.gardev@akros.ch>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* kv-cache : simplify the "struct llama_kv_cache" interface
ggml-ci
* kv-cache : revert the (n_swa + n_ubatch) change (for next PR)
ggml-ci
* kv-cache : some comments
ggml-ci
* context : fix graph reserve for multiple sequences
ggml-ci
* kv-cache : fix typo [no ci]
* kv-cache : fix find_slot() logic for free slots
ggml-ci
* llama : add TODO for deprecating the defrag API in the future
* kv-cache : improve find_slot() using min/max seq pos info
ggml-ci
* llama : handle aborts and compute errors
ggml-ci
* memory : extract state into llama_memory_state
ggml-ci
* kv-cache : add comments
ggml-ci
* server : update batching logic to reset n_batch on successful decode
* server : upon full re-processing, remove the sequence from the cache
* kv-cache : add TODO for doing split_equal when split_simple fails
ggml-ci
* convert: add support for BertForSequenceClassification
* add support for reranking using BertForSequenceClassification
* merge checks of eos and sep
* fix lint
---------
Co-authored-by: dinhhuy <huy.dinh@brains-tech.co.jp>
* mtmd : allow multiple modalities at the same time
* refactor mtmd tokenizer
* fix compile
* ok, missing SinusoidsPositionEmbedding
* first working version
* fix style
* more strict validate of n_embd
* refactor if..else to switch
* fix regression
* add test for 3B
* update docs
* fix tokenizing with add_special
* add more tests
* fix test case "huge"
* rm redundant code
* set_position_mrope_1d rm n_tokens
* add preludes to content on partial regex match
* allow all parsers to parse non-tool-call content.
* tweak order of <|python_tag|> vs <function= parsing for functionary v3.1 format. still not ideal but hopefully less prone to crash
* fix deltas of tool_call.function.name
* fix tool_call.id (was in tool_call.function.id!) + add function type
* add tool_call.type
* populate empty tool_call.function.arguments on first delta
* convert ok, load ok
* warmup ok
* test
* still does not work?
* fix padding
* temporary give up
* fix merge conflict
* build_ultravox()
* rm test
* fix merge conflict
* add necessary mtmd APIs
* first working version (only 4s of audio)
* will this monster compile?
* fix compile
* please compile
* fPIC
* fix windows
* various fixes
* clean up audio_helpers
* fix conversion
* add some debug stuff
* long audio input ok
* adapt the api
* add --audio arg
* final touch UX
* add miniaudio to readme
* fix typo
* refactor kv metadata
* mtmd_default_marker()
* Add the endpoints /api/tags and /api/chat
Add the endpoints /api/tags and /api/chat, and improved the model metadata response
* Remove trailing whitespaces
* Removed code that is not needed for copilot to work.
* server : fix first message identification
When using the OpenAI SDK (https://github.com/openai/openai-node/blob/master/src/lib/ChatCompletionStream.ts#L623-L626) we noticed that the expected assistant role is missing in the first streaming message. Fix this by correctly checking for the first message.
Co-authored-by: Piotr Stankiewicz <piotr.stankiewicz@docker.com>
Signed-off-by: Dorin Geman <dorin.geman@docker.com>
* server : Fix checks for first role message for stream=True
Co-authored-by: Piotr Stankiewicz <piotr.stankiewicz@docker.com>
Signed-off-by: Dorin Geman <dorin.geman@docker.com>
---------
Signed-off-by: Dorin Geman <dorin.geman@docker.com>
Co-authored-by: Piotr Stankiewicz <piotr.stankiewicz@docker.com>
* Remove mmap workaround on windows
After some testing I found that mmap is supported on windows and for
many GPUs on Linux. Therefore I remove the workaround for windows since
it is not necessary.
* Update llama-bench README
SYCL backend introduced a workaround that allows execution of
llama-bench also without specifying `--mmp 0` flag
* webui : improve accessibility for visually impaired people
* add a11y for extra contents
* fix some labels being read twice
* add skip to main content
* server: Allow pasting file from clipboard
* server: Prevent default action on file paste
* update build
* format then build combined
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* convert : internvl support
* InternVL3-1B working
* fix regression
* rm mobilevlm from test
* fix conversion
* add test for internvl
* add to list of pre-quant
* restore boi/eoi check
* add clarify comment for norm eps
* server : (experimental) vision support via libmtmd
* mtmd : add more api around mtmd_image_tokens
* mtmd : add more api around mtmd_image_tokens
* mtmd : ability to calc image hash
* shared_ptr for mtmd_image_tokens
* move hash to user-define ID (fixed)
* abstract out the batch management
* small fix
* refactor logic adding tokens to batch
* implement hashing image
* use FNV hash, now hash bitmap instead of file data
* allow decoding image embedding to be split into batches
* rm whitespace
* disable some features when mtmd is on
* fix --no-mmproj-offload
* mtmd_context_params no timings
* refactor server_inp to server_tokens
* fix the failing test case
* init
* wip
* working version
* add mtmd::bitmaps
* add test target
* rm redundant define
* test: mtmd_input_chunks_free
* rm outdated comment
* fix merging issue
* explicitly create mtmd::input_chunks
* mtmd_input_chunk_copy
* add clone()
* improve server_input struct
* clip : fix confused naming ffn_up and ffn_down
* rm ffn_i/o/g naming
* rename n_embd, n_ff
* small fix
* no check n_ff
* fix detokenize
* add const to various places
* add warning about breaking changes
* add c api
* helper: use mtmd_image_tokens_get_n_pos
* fix ctx_shift
* fix name shadowing
* more strict condition
* support remote image_url
* remote image_url log
* add CI test
* do not log base64
* add "has_multimodal" to /props
* remove dangling image
* speculative: use slot.cache_tokens.insert
* Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* rm can_be_detokenized
* on prmpt processing done, assert cache_tokens.size
* handle_completions_impl returns void
* adapt the new web ui
* update docs and hot topics
* rm assert
* small fix (2)
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* rework the input area
* process selected file
* change all icons to heroicons
* fix thought process collapse
* move conversation more menu to sidebar
* sun icon --> moon icon
* rm default system message
* stricter upload file check, only allow image if server has mtmd
* build it
* add renaming
* better autoscroll
* build
* add conversation group
* fix scroll
* extra context first, then user input in the end
* fix <hr> tag
* clean up a bit
* build
* add mb-3 for <pre>
* throttle adjustTextareaHeight to make it less laggy
* (nits) missing padding in sidebar
* rm stray console log
* mtmd : refactor graph builder
* fix qwen2vl
* clean up siglip cgraph
* pixtral migrated
* move minicpmv to a dedicated build function
* move max_feature_layer to build_llava
* use build_attn for minicpm resampler
* fix windows build
* add comment for batch_size
* also support tinygemma3 test model
* qwen2vl does not use RMS norm
* fix qwen2vl norm (2)
Alex Wu
compilade
Sigbjørn Skjæret
R0CKSTAR
Johannes Gäßler
tc-mb
g2mt
Lukas Straub
Ed Addario
Daniel Bevenius
Radoslav Gerganov
Xuan-Son Nguyen
kiwi
stduhpf
Molly Sophia
IsaacDynamo
Georgi Gerganov
Douglas Hanley
Eric Zhang
Alawode Oluwandabira
Vedran Miletić
matteo
Renat
Nigel Bosch
Ruikai Peng
yuiseki
aa956
bashayer hijji
Eric Curtin
Aman
Taylor
Juk Armstrong
Olivier Chafik
Max Krasnyansky
igardev
Đinh Trọng Huy
Sky
Percy Piper
Robin Davidsson
Dorin-Andrei Geman
l3utterfly
Nicolò Scipione
Isaac McFadyen
Diego Devesa
Piotr Wilkin (ilintar)
Olivier Chafik
Luca Stefani
Anudit Nagar
City
Anthony Umfer
David Huang
Bartowski
R0CKSTAR
Matt Clayton
welix
oobabooga