- Change '--embedding' to '--embeddings' in the README
- Update the description to match the latest --help output
- Added a caution about defining physical batch size
* [server] Cleanup a memory leak on exit
There are a couple memory leaks on exit of the server. This hides others.
After cleaning this up, you can see leaks on slots. But that is another
patch to be sent after this.
* make tab into spaces
* feat: first things to do
* feat: create tensors for Jina architecture
* fix: use other tensors
* feat: embedding gets results
* fix: fix usage of ALIBI
* fix: clean prints
* fix: do some cleanup unused vars
* fix: revert changes to Makefile and CMakeLists
* fix: revert some changes
* fix: fix small detail
* fix: fix convert formatting
* fix: fix linting and editor
* feat: set proper vocab settings
* fix: JinaBertForMaskedLM registration
* feat: support q_normalization and k_normalization in Jina arch
* feat: handle gpt2 tokenizer with Jina architecture
* feat: example comments in embedding
* feat: rename Jina Bert to Jina Bert V2
* fix: add some changes as per review
* feat: proper KQ_pos for Jina embeddings
* feat: add capacity to load models ES and DE for Spanish
* llama : fix pre-tokenizers
* ggml : full ALiBi support
* ggml : update ggml_soft_max_ext() CUDA, SYCL
* ggml : ggml_flash_attn_ext() support ALiBi (CPU)
* ggml : ggml_flash_attn_ext() support ALiBi (Metal)
* ggml : fix warning
* ggml : ggml_flash_attn_ext() support ALiBi (CUDA)
ggml-ci
* minor : clean-up
* embedding : add warning about missing SEP
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
The llama.cpp grammar parser had a bug where forgetting to add a closing
quotation mark to strings would cause parsing to crash. Anyone running a
server on a public endpoint is advised to upgrade. To reproduce this bug
./llamafile -m foo.gguf -p bar --grammar 'root::="'
Credit for discovering and reporting this issue goes to Eclypsium
Security Researcher Richard Johnson <Richard.johnson@eclypsium.com>.
* Revert "Revert "llava : add support for moondream vision language model (#6899)""
This reverts commit 9da243b36a.
* Fix num_positions and embeddings initialization
* convert-hf : begin refactoring write_tensor
* convert : upgrade to sentencepiece v0.2.0
* convert-hf : remove unused n_dims in extra_*_tensors
* convert-hf : simplify MoE weights stacking
* convert-hf : flake8 linter doesn't like semicolons
* convert-hf : allow unusual model part names
For example, loading `model-00001-of-00001.safetensors` now works.
* convert-hf : fix stacking MoE expert tensors
`torch.stack` and `torch.cat` don't do the same thing.
* convert-hf : fix Mamba conversion
Tested to work even with a SentencePiece-based tokenizer.
* convert : use a string for the SentencePiece tokenizer path
* convert-hf : display tensor shape
* convert-hf : convert norms to f32 by default
* convert-hf : sort model part names
`os.listdir` is said to list files in arbitrary order.
Sorting the file names should let "model-00009-of-00042.safetensors"
be loaded before "model-00010-of-00042.safetensors".
* convert-hf : use an ABC for Model again
It seems Protocol can't be used as a statically type-checked ABC,
because its subclasses also can't be instantiated. (why did it seem to work?)
At least there's still a way to throw an error when forgetting to define
the `model_arch` property of any registered Model subclasses.
* convert-hf : use a plain class for Model, and forbid direct instantiation
There are no abstract methods used anyway,
so using ABC isn't really necessary.
* convert-hf : more consistent formatting of cmdline args
* convert-hf : align the message logged for converted tensors
* convert-hf : fix Refact conversion
* convert-hf : save memory with lazy evaluation
* convert-hf : flake8 doesn't like lowercase L as a variable name
* convert-hf : remove einops requirement for InternLM2
* convert-hf : faster model parts loading
Instead of pre-loading them all into a dict, iterate on the tensors
in the model parts progressively as needed in Model.write_tensors
Conversion for some architectures relies on checking for the presence
of specific tensor names, so for multi-part models, the weight map is read
from the relevant json file to quickly get these names up-front.
* convert-hf : minor changes for consistency
* gguf-py : add tqdm as a dependency
It's small, and used for a progress bar
in GGUFWriter.write_tensors_to_file
* Added themes support with two sample themes and a favicon.
* Newline
* Newline
* Newline
* Trailing whitespace
* Increased opacity for contrast
* Increase opacity.
Check actions cancelled for some other priority job and I can't seem to manually re-run them, so MOAR OPACITY
* Opacity action trigger.
Trying to re-trigger the cancelled action.
* One more opacity adjustment
This Actions pipeline is failing for random issues.
* Delete examples/server/themes/buttons_top/completion.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/buttons_top/index.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/completion.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/buttons_top/json-schema-to-grammar.mjs
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/index.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/json-schema-to-grammar.mjs
This will be served from the static string built-in to server.
* Replaced underscore.
* Introduce bfloat16 support
Many models on Hugging Face (e.g. Mistral, TinyLLaMA) use bfloat16 as
their canonical floating point format.
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌──┴───┐┌─┴───┐
0b0000000000000000 brain16
This encoding has the same number of exponent bits as float32. That
makes conversion relatively straightforward, even in the absence of
hardware support. For example, converting brain16 to binary32 means
simply shifting 16 bits to the left.
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌──┴───┐┌─┴───────────────────┐
0b00000000000000000000000000000000 IEEE binary32
The issue is that converting bf16 to fp16 can result in information
loss. Only 13% of bf16 numbers can be precisely represented in fp16
which in practice ends up being 99.71% of Mistral 7b v0.2's weights
however there is currently no way other than fp32 to get the others
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌─┴─┐┌─┴──────┐
0b0000000000000000 IEEE binary16
This change fixes that, by adding a bf16 data type to GGML. Support
for CPU inference has been implemented along with optimizations for
the AVX2, AVX512, and AVX512BF16 ISAs. Perplexity on Mistral 7b 0.2
improves somewhere around -0.0024 to -0.0046 compared to using fp16
* Remove GGML code that's not needed
* Minimize the GGML API surface area for BF16
* Remove bf16 luts
* Make the GGML header look nicer
* Fix documentation
* Apply ggerganov's fixes for test-backend-ops
* Add BF16 code for new ggml_validate_row_data() function
* Fixed save_imatrix to match old behaviour for MoE
This fix is simple and clear, but unnecessarily doubles the memory overhead..
* Fixed missing idx variable
* Unconditionally increment ncall
Co-authored-by: slaren <slarengh@gmail.com>
* Fixed 2 bugs in save_imatrix()
- Fixed segfault bug because the counts vector needed to be created.
- Fixed pre-existing bug didn't actually add to the counts for "--combine" option.
* ncall needs summing too
* Trailing whitespace
---------
Co-authored-by: slaren <slarengh@gmail.com>
* Update log text (EOS to EOG)
The log text "found EOS" is no longer always correct, here, because there is now an is-EOG check that also returns true for EOT.
* Improve log msg. further by using "an" instead of "some".
As suggested, to avoid misunderstanding (no multiple EOG tokens found, just one).
This will reproduce the issue in llama13b
{
'prompt': 'Q: hello world \nA: ',
'stop': ['\n'],
'temperature': 0.0,
'n_predict': 10,
'cache_prompt': True,
'n_probs': 10
}
* ggml : add ggml_flash_attn_ext API
* ggml : fix GQA support in ggml_flash_attn_ext
* ggml : online attention (CPU)
* metal : initial implementation
* metal : f16 precision
* metal : reduce branches
* metal : specialize for head size
* wip : 8 rows per simd group
* wip : 4 rows per simd group
* wip : template for rows per warp
* metal : parallelize across KV size
* metal : parallel reduce across heads
* metal : efficient flash_attn_f16 implementation
* metal : avoid redundant loads of the attention
* metal : scale and mask in matrix form
* metal : fix comment
* llama : avoid ggml_cast, use F32 query
* metal : add parallel reduce version (disabled)
* metal : move output into local memory + optimize
- the result from each simdgroup now stays in the registers
- significantly reduced SRAM usage
- more efficient skipping of -INF blocks
- avoid simdgroup barrier in hot loop
- add comments
* metal : add tests, fix scaling, support C > 32
* metal : improve precision
* ggml : fix f16 mad
* metal : minor
* metal : support Q > 8
* tests : add ATTN tests
* metal : disable buffer allocation logs
* tests : more
* metal : faster inner loop for C == 32
* metal : fix array initialization
* tests : ifdef
* ggml : switch to padded F16 mask for ggml_soft_max, ggml_flash_attn_ext
* ggml : fix ggml_soft_max mask requirement
* cuda : fix soft_max to use correct mask size
* cuda : add flash_attn kernel (wip)
* metal : optimize softmax for C > 32
* metal : optimize softmax
* tests : minor fix
* cuda : avoid zeroing fragments
* tests : update dims
* cuda : fix __hisinf() result check
* cuda : avoid warp_reduce for smax
* cuda : use int instead of int64_t
Noticeably improves performance (thanks to Johannes)
* cuda : make loops use the same loop values
Thanks Johannes again for the tip
* cuda : unroll some of the loops
* cuda : avoid __hisinf branches
* cuda : use half2 in softmax
* cuda : switch to 1 warp for bs > 16
* cuda : speed-up reduce part of the kernel
* cuda : unroll Q*K^T loop
* cuda : fix -INF block check
* cuda : simplify softmax
* cuda : fix matrix names
* cuda : minor
* llama : adapt to F16 KQ_pos
* llama : adapt new models to F16 KQ_mask
* ggml : fix F16 store (ARM NEON)
* llama : fix type of KQ_mask and KQ_pos
* ggml : fix CPU soft_max
* tests : add hs=256
* cuda : fix build
* metal : improve perf via smaller int registers
* cuda : adapt soft_max to F16 mask and pos
* CUDA: faster FlashAttention, kernel for bs == 1
* 16 cols for Phi-2
* no vec for hs, no hs==256 ncols==32 for Volta
* adjust kernel selection logic
* 4 warps, 256 stride for all D
* no ncols == 64
* Multiple parallel blocks for batch size 1
* fix compile warnings
* fix excessive KQ_b loads
* fix cmake build
* fix KV cache padding, NaN from INFINITY (#6438)
* llama : flash_attn cparam + fix defrag
* server: support flash_attn param
* server: bench: enable flash_attn param
* CUDA: refactor host code, dyn. par. blocks
* fix flash_attn_vec_f16 race condition
* flush softmax exp below threshold to 0
* store temp KQ in registers
* Calculate KQ as FP32 if KQV has GGML_PREC_F32
* Add __hgt2_mask implementation for CUDA 11
* fix KQ FP32 precision fpr parallel_blocks > 1
* llama-bench : add -fa,--flash-attn arg
* metal : add BS=1 kernel for flash attention (#6508)
* metal : add BS=1 kernel for flash attention (wip)
* metal : support more than 1 warps
* metal : opts
* metal : opt
* metal : switch to parallel reduce
* metal : reduce registers
* metal : simplify
* metal : initial FA vec kernel
* metal : use F32 attention accumulators
* batched-bench : add fattn arg
* llama : simplify llama_build_kv_store
ggml-ci
* llama : adapt build_olmo to changes
* ggml : fix arm fp16 store on windows
* metal : clean-up
* metal : clean-up kernel code
* metal : minor
* tests : remove benchmarks
ggml-ci
* ggml : fix avx512 const correctness
ggml-ci
* ggml : fix soft_max with bias on CPU
ggml-ci
* common : print --flash-attn in help
* ggml : fix num dimensions in ggml_flash_attn_ext
* llama : force disable flash attention for incompatible models
* ggml : ggml_soft_max support F16/F32 mask/pos
ggml-ci
* cuda : uint -> uint32_t
* cuda : "constexpr dim3" -> "const dim3"
ggml-ci
* cuda : try to fix __hgt2_mask
ggml-ci
* ggml : add TODO's for F16/F32 mask/pos support in other backends
* llama : replace bool need_kq_pos with use_alibi
* llama : prep ALiBi support for BERT models
ggml-ci
* llama : fix n_batch requirements
ggml-ci
* cont
* server : add help for --flash-attn arg
* llama : disable FA for AMD
* tests : remove TMP_ATTN_BENCH
ggml-ci
* llama : support save/load state with FA enabled
ggml-ci
* ci : add CUDA save-load-state tests
ggml-ci
* llama : llama_kv_cache_clear zeroes data + fix save-load seq
ggml-ci
* llama : fix copy-paste errors, add TODO
* llama : disallow incompatible states
* llama : update llama_state_get_size after v_trans field
* metal : remove tmp log
* llama : add static reminder for llama_state_get_size
* metal : fix max nsg
ggml-ci
* ci : fix arg order
ggml-ci
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
Co-authored-by: Pierrick HYMBERT <pierrick.hymbert@gmail.com>
* imatrix: save the dataset file used in the output file
* llama: support kv overrides type string string
* common: factorize KV Overrides parsing between common and server
* quantize: add imatrix n entries and dataset KV metadata
quantize: factorize KV Overrides parsing between common
#6656
* llama: remove kv override str_value initialization as it does not compile on some toolchain
* quantize: add imatrix m_last_call as `quantize.imatrix.chunks_count`
* quantize: add imatrix filename in KV
* llama: add llama_model_kv_override_free
* common: add llama_model_kv_override_free
common: free kv override if used after model loading
* llama: finally move the string KV override value to the stack
* llama : minor
* no need to add a NUL to the std::vector, std::string can be initialized from a pair of iterators.
Co-authored-by: slaren <slarengh@gmail.com>
* kv override: ensure string termination
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: slaren <slarengh@gmail.com>
* server: cap n_predict if not set to n_ctx_train
* server: fix infinite loop
* server: infinite loop, move in process_token
server: infinite loop: set stop limit to true
* minor: spaces
* minor: spaces
* server: include prompt tokens in the EOS limit
* add support for moondream vision language model
This required making the following changes to the CLIP model:
1. Support for patch embedding bias.
2. Make class embedding and pre-layernorm optional.
3. Add support for post-layernorm.
* Update examples/llava/clip.cpp
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit renamesthe lerp (linear interpolation) function in clip.cpp
to avoid a conflict with the lerp function in the <cmath> standard C++
library when using c++20.
The motivation for this change is to enable projects that use c++20 to
be able to compile clip.cpp without having to resort to patching it. The
lerp function was added to cmath in version C++20 (202002L) and is why
this is not causing any issue at the moment as C++11/C++17 is currently
used by llama.cpp.
I realize that llama.cpp uses either C++11 (or C++17 in the case for
SYCL) but wanted to ask if this would be an acceptable change just the
same.
Refs: https://en.cppreference.com/w/cpp/numeric/lerp
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* Implement '--keep-split' to quantize model into several shards
* Add test script
* Update examples/quantize/quantize.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Split model correctly even if tensor id is out-of-order
* Update llama_model_quantize_params
* Fix preci failures
---------
Co-authored-by: z5269887 <z5269887@unsw.edu.au>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* fix: revert showing control tokens by default
* feat: revert changes to default behavior of llama_token_to_piece; provide overridden declaration to receive "bool special" param to toggle showing control tokens
* feat: use the overridden declaration of llama_token_to_piece from common/common.cpp to specify "false" so that control tokens are not shown in chat completion responses"
* common : simplify
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* `build`: generate hex dumps of server assets on the fly
* build: workaround lack of -n on gnu xxd
* build: don't use xxd in cmake
* build: don't call xxd from build.zig
* build: more idiomatic hexing
* build: don't use xxd in Makefile (od hackery instead)
* build: avoid exceeding max cmd line limit in makefile hex dump
* build: hex dump assets at cmake build time (not config time)
* Support Llama 3 conversion
The tokenizer is BPE.
* style
* Accept suggestion
Co-authored-by: Sourab Mangrulkar <13534540+pacman100@users.noreply.github.com>
* llama : add llama_token_is_eog()
ggml-ci
* llama : auto-detect more EOT tokens when missing in KV data
* convert : replacing EOS token is a hack
* llama : fix codegemma EOT token + add TODOs
* llama : fix model type string for 8B model
---------
Co-authored-by: Sourab Mangrulkar <13534540+pacman100@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* ggml : group all experts in a single ggml_mul_mat_id
cuda : improve mmid row copy
* cuda : fix bin bcast with non-cont src0
* test-backend-ops : only run all mul mat tests for base types
* llama : disable moe offloading with SYCL
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This change upstreams llamafile's cpu matrix multiplication kernels
which improve image and prompt evaluation speed. For starters, Q4_0
and Q8_0 weights should go ~40% faster on CPU. The biggest benefits
are with data types like f16 / f32, which process prompts 2x faster
thus making them faster than quantized data types for prompt evals.
This change also introduces bona fide AVX512 support since tinyBLAS
is able to exploit the larger register file. For example, on my CPU
llama.cpp llava-cli processes an image prompt at 305 tokens/second,
using the Q4_K and Q4_0 types, which has always been faster than if
we used f16 LLaVA weights, which at HEAD go 188 tokens/second. With
this change, f16 LLaVA performance leap frogs to 464 tokens/second.
On Intel Core i9-14900K this change improves F16 prompt perf by 5x.
For example, using llama.cpp at HEAD with Mistral 7b f16 to process
a 215 token prompt will go 13 tok/sec. This change has fixes making
it go 52 tok/sec. It's mostly thanks to my vectorized outer product
kernels but also because I added support for correctly counting the
number of cores on Alderlake, so the default thread count discounts
Intel's new efficiency cores. Only Linux right now can count cores.
This work was sponsored by Mozilla who's given permission to change
the license of this code from Apache 2.0 to MIT. To read more about
what's improved, and how it works, see: https://justine.lol/matmul/
This commit updates the hf.sh script usage to include the --outdir option
and specifies the models directory as the output directory.
The motivation for this is to avoid cluttering the root directory with
model files.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* Fix --split-max-size
Byte size calculation was done on int and overflowed.
* add tests.sh
* add examples test scripts to ci run
Will autodiscover examples/*/tests.sh scripts and run them.
* move WORK_PATH to a subdirectory
* clean up before and after test
* explicitly define which scripts to run
* add --split-max-size to readme
* disable mmap to fix memcpy crash, add missed cmd in guide, fix softmax
* refactor to disable mmap for SYCL backend
* fix compile error in other os
* refactor the solution, use host buf to fix it, instead of disable mmap
* keep to support mmap()
* use host buff to reduce malloc times
* revert to malloc/free solution, for threaad safe
* infill : add download instructions for model
This commit adds instructions on how to download a CodeLlama model
using the `hf.sh` script. This will download the model and place it
in the `models` directory which is the same model use later by the
infill example.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* squash! infill : add download instructions for model
Clarify the reason for using CodeLlama.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
---------
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* gguf-debug: Example how to use ggml callback for debugging
* gguf-debug: no mutex, verify type, fix stride.
* llama: cv eval: move cb eval field in common gpt_params
* ggml_debug: use common gpt_params to pass cb eval.
Fix get tensor SIGV random.
* ggml_debug: ci: add tests
* ggml_debug: EOL in CMakeLists.txt
* ggml_debug: Remove unused param n_batch, no batching here
* ggml_debug: fix trailing spaces
* ggml_debug: fix trailing spaces
* common: fix cb_eval and user data not initialized
* ci: build revert label
* ggml_debug: add main test label
* doc: add a model: add a link to ggml-debug
* ggml-debug: add to make toolchain
* ggml-debug: tests add the main label
* ggml-debug: ci add test curl label
* common: allow the warmup to be disabled in llama_init_from_gpt_params
* ci: add curl test
* ggml-debug: better tensor type support
* gitignore : ggml-debug
* ggml-debug: printing also the sum of each tensor
* ggml-debug: remove block size
* eval-callback: renamed from ggml-debug
* eval-callback: fix make toolchain
---------
Co-authored-by: slaren <slarengh@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit adds an option to the gguf example to not check the tensor
data.
The motivation for this is that it can be nice to use the gguf tool to
read other .gguf files that were not created by the gguf tool.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
Key changes:
* BERT conversion: fix abuse of LlamaHfVocab, do not set BOS or EOS
* Nomic Embed conversion: pad vocab instead of slicing embedding tensor
* llama_tokenize: handle added special tokens like HF does
* llama : save and restore kv cache for single seq id
* remove trailing whitespace
* respond error in case there's no space in the kv cache
* add kv seq save restore to test case
* add --slot-save-path arg to enable save restore and restrict save location
* Returning 0 for some cases, instead of asserting.
* cleanup error cases
* rename sequence state functions
* rename state get set functions
* add previous function names back in with DEPRECATED notice
* update doc
* adjust endpoints to preferred style
* fix restoring zero cell count
* handle seq rm return value
* unused param
* keep in the size check
* fix return types
* add server test case for slot save restore
* cleanup
* add cake
* cleanup style
* add special
* removing a whole sequence never fails
* move sequence state file functionality from server to llama to match session api and add version tags
* catch exceptions on save as well
* error log messages
* check types for stricter restore
* update server doc
* readme : update API changes date
* strict filename validation
* move include, reject bom as well
* also reject empty filename
* reject whitespace and trailing dot
---------
Co-authored-by: Martin Evans <martindevans@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* ci: bench: support sse and fix prompt processing time
server: add tokens usage in stream mode
* ci: bench: README.md EOL
* ci: bench: remove total pp and tg as it is not accurate
* ci: bench: fix case when there is no token generated
* ci: bench: change to the 95 percentile for pp and tg as it is closer to what the server exports in metrics
* ci: bench: fix finish reason rate
* ci: bench: change trigger path to not spawn on each PR
* ci: bench: add more file type for phi-2: q8_0 and f16.
- do not show the comment by default
* ci: bench: add seed parameter in k6 script
* ci: bench: artefact name perf job
* Add iteration in the commit status, reduce again the autocomment
* ci: bench: add per slot metric in the commit status
* Fix trailing spaces
* Typo fix to server's README.md
Fix minor typo ("tonen") in server README.
* server readme grammar/style fixes.
Quickly went through this file to look for inconsistencies in
presentation of defaults, flag options, and looked for typos
and grammar issues.
Not perfect, but hopefully improved.
* Update README.md
Remove an extra space before newline.
* ggml : update mul_mat_id to use the same tensor for all the experts
* update cuda
* minor
* update metal
* update test-backend-ops
* fix cuda
* Update ggml-metal.m
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* update convert.py
* update convert-hf-to-gguf.py
* update convert.py for mixtral hf models
* Update convert-hf-to-gguf.py
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* cuda : support non-pow-2 number of experts
* allow quantize to work for split and merged experts models in the same way
* cleanup + disable mmap automatically with split tensors models
* update imatrix
* test-backend-ops : test qwen argsort
* update grok model loading
* llama : add merged experts tensors to the grok tensor map
* minor
* gguf : bump version
* fix quantizing of merged experts
* convert-hf-to-gguf.py : update grok (untested)
* make linter happy
* cuda/argsort : use shared memory instead of pool memory
* convert : fix grok tensor names
* metal : add support for non-pow-2 argsort
* llama : more loader cleanup, better error checking
* cuda : fix warning
* llama : still use mmap for loading old models, but copy the data to a host buffer
* add review note
* llama : remove ffn tensor counting + add sanity check
ggml-ci
* convert : fix handling of n_experts == None
ggml-ci
* imatrix : fix ncall counters
* llama : produce error if imatrix size does not match
* quantize : terminate on errors + trace logs
ggml-ci
* metal : pad shared memory to 16 bytes
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* embedding : show full embedding for single prompt
To support the use case of creating an embedding for a given prompt, the entire embedding and not just the first part needed to be printed.
Also, show cosine similarity matrix only if there is more than one prompt, as the cosine similarity matrix for a single prompt is always `1.00`.
* Update examples/embedding/embedding.cpp
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* llama : greatly reduce logits memory usage
* llama : more compact state saving and reloading
* llama : fix lctx.n_outputs not being set before building graph
* perplexity : adapt to the logits API changes
* perplexity : fix Winogrande, use correct logits for second choice start
The first logits used to evaluate the second choice were not from
the end of the common prefix; instead, they were the logits from the end
of the first choice. This has been corrected.
The previous implementation sometimes had outliers in the scores of
choices for some tasks, and the logic to skip choices words
in the log-likelihood evaluation probably was an attempt to reduce those,
but it was complex and didn't quite seem to be the right thing.
This is simpler now, and the outlier scores aren't there anymore.
* perplexity : normalize spaces and punctuation in Winogrande sentences
* llama : fix embedding conditions
* llama : fix llama_get_embeddings_ith when the resulting id is 0
* llama : fix wrong n_outputs in llama_set_inputs
A mismatch happened when using a smaller n_ubatch than n_batch and then using
llama_batch_get_one(). The decision of what n_outputs should be now almost
fully depends on how lctx.n_outputs is set in llama_decode_internal.
The conditions are simpler this way.
* llama : when saving the state, recalculate n_outputs
This ensures the correct number of outputs for the entire previous batch
is stored in the session file, even when n_ubatch is smaller than n_batch.
* llama : fix not-skipping outputs of non-causal models
* llama : fix running a batch with n_outputs == 0
It previously worked because lctx.inp_out_ids was not initialized,
so it pointed to some garbage address which was somehow still valid when I
ran my tests.
* llama : keep same graph topology even when n_outputs == 0
* ggml : saner ggml_can_repeat with empty tensors
* ggml : future-proof ggml_is_empty by using GGML_MAX_DIMS - 1
* ggml : do not multi-thread ops returning empty tensors
* ggml : make ggml_is_empty public and work with views
* llama : use a vector for ctx->output_ids
* llama : rework reallocation logic for llama_output_reserve
Now comparing the actual size with the new total size of the output buffer
to allow more efficient enabling and disabling of the embeddings
and/or logits output in the future.
* ggml : skip empty tensors in all backends
* llama : fix llama_output_reserve nullptr deref when new_size is 0
* perplexity : make Winogrande work as it does on master
The problems with the Winogrande implementation will
need to be fixed in a separate PR to ease review.
* llama : clearer error messages for invalid logits or embeddings ids
* llama : assert all models that can have inp_out_ids
Since the graph topology is now constant, this presence check
can be done even when there are no outputs.
* llama : assert logits and embd buffers exist before writing to them
* llama : handle errors from llama_output_reserve at call sites
* perplexity : make hellaswag and multiple-choice outputs identical to master
Due to how the KV cache is updated, the logprobs for tokens in a batch
are very slightly affected by the other tokens present in the batch,
so to make hellaswag and multiple-choice return exactly the same results
as on master, the last token of each sequence needs to be evaluated
even though its output is not used at all.
This will probably be changed back in the future to make these benchmarks
a tiny bit faster.
* perplexity : fix division by zero when using less than 100 multiple-choice tasks
* llama : allow loading state saved with a different ctx size
When loading a session file, the context size is now only required to be
at least enough to load the KV cells contained in that session file,
instead of requiring to use exactly the same context size as when saving.
Doing this enables the use-case of extending or shrinking the context size
of a saved session.
This breaks existing session files because the meaning of kv_buf_size
is slightly changed (previously it was the size of the whole KV cache,
now it's only the size of the saved part of it). This allows for
finer-grained sanity checks when loading in an effort to keep kv_buf_size
useful even when the kv_size is changed.
* llama : minor
ggml-ci
* readme : update recent API changes, and warn about Vulkan
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* iq1_m: basics
* iq1_m: basics-2
* iq1_m: CUDA dequantize works
Very 1st shot I get PPL = 9.76 for LLaMA-v2-7B.
* iq1_m: separate shifts for each group of 8 in a block
We get
PPL(LLaMA-v2-7B ) = 9.2810
PPL(LLaMA-v2-13B) = 6.8105
Not bad, but slightly higher than
sqrt(PPL(IQ1_S) * PPL(IQ2_XXS))
which is the expected outcome given that IQ1_M is
halfway between IQ1_S and IQ2_XXS in terms of bpw.
From this, we would expect
PPL = 9.14 for LLaMA-v2-7B
PPL = 6.63 for LLaMA-v2-13B
* iq1_m: go to 3-bit scales
There is slight increase in PPL, but the 0.0625 bpw reduction
in size is totally worth it.
We now have
PPL(LLaMA-v2-7B ) = 9.4469 at 1.96 bpw
PPL(LLaMA-v2-13B) = 6.8717 at 1.93 bpw
PPL(LLaMA-v2-70B) = 4.8568 at 1.85 bpw
* iq1_m: scalar dot product
* iq1_m: AVX2 dot product
* iq1_m: very slightly faster AVX2 dot product
* iq1_m: ARM_NEON dot product
Works, but very slow (10.5 t/s)
* iq1_m: Metal - dequantize works, dot product does not
* iq1_m: Metal now works
About the same performance as iq1_s.
* iq1_m: minor
* iq1_m: checking pure iq1_m quantization
It is pretty bad: PPL(LLaMA-v2-7B) = 34 if we quantize output.weight
with Q4_K.
* iiq1_m: slightly faster ARM_NEON dot product
10.5 t/s -> 11.65 t/s
* iq1_m: faster ARM_NEON dot product
11.65 t/s -> 14.9 t/s
* iq1_m: another minor ARM_NEON dot product improvement
14.9 -> 15.0 t/s
* iq1_m: small PPL improvement via super-block scale adjustment
After quantizing block scales redo the super-block scale fit.
PPL(LLaMA-v2-7B ) = 9.3346
PPL(LLaMA-v2-13B) = 6.8419
PPL(LLaMA-v2-70B) = 4.8294
PPL(Mistral-7B ) = 8.1624
* iq1_m: adapt to CUDA refactoring
* iq1_m: remove unused variable
We have progressed to warnings being errors.
* iq1_m: add to backend-ops tests
* iq1_m: fix Windows ARM
* iq1_m: use common definition of iq1m_scale_t
* cuda: assert -> NO_DEVICE_CODE
* iq1_M: PR comments
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
* quantize: be able to override metadata by key
* minor : spacing
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* embedding: assign `n_ubatch` value, print error on `n_batch` overflow
* Update examples/embedding/embedding.cpp
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* use %ld instead of %lld
* Revert "use %ld instead of %lld"
This reverts commit ea753ede90.
---------
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* sampling: remove duplicated code for probability distribution access
* free original_logits
* fix original_logits allocation
* fixes based on review @cebtenzzre
* change function name to `llama_sampling_prepare`
* llama: llama_split_prefix fix strncpy does not include string termination
common: llama_load_model_from_url:
- fix header name case sensitive
- support downloading additional split in parallel
- hide password in url
* common: EOL EOF
* common: remove redundant LLAMA_CURL_MAX_PATH_LENGTH definition
* common: change max url max length
* common: minor comment
* server: support HF URL options
* llama: llama_model_loader fix log
* common: use a constant for max url length
* common: clean up curl if file cannot be loaded in gguf
* server: tests: add split tests, and HF options params
* common: move llama_download_hide_password_in_url inside llama_download_file as a lambda
* server: tests: enable back Release test on PR
* spacing
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* spacing
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* spacing
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* quantize: be able to specify the output tensor type
* quantize: be able to specify the token embedding tensor type
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
* split: support in llama_model_loader
* avoid copying the entire vector
Co-authored-by: slaren <slarengh@gmail.com>
* split: move llama_tensor_offset to llama_model_loader
* llama_model_loader: PR feedbacks:
- use only one gguf_context for metadata only
- store all ggml_context in a vector as the files and mappings
- store all weights in a vector along with the source tensor
- rename ctx_gguf to meta
- rename ctx_meta to contexts
* avoid copying the entire vector
* Simplify this by making these optional, switch some layer creation tensor optional
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Handle optional tensors
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* llama_model_loader: fail if backend cannot allocate buffer
* fix mmap buffer management
* llama_model_loader: map file to backend buffer if the allocation succeeds only
* llama_model_loader: only map tensors included in the context
* llama_model_loader: minor, use same variable name for consistency, fix spacing in types cast
* llama_model_loader: fail if any of backend buffer cannot be allocated
* spacing
Co-authored-by: slaren <slarengh@gmail.com>
* fix loop over pointer
Co-authored-by: slaren <slarengh@gmail.com>
* llama_model_loader: if n_tensors declared not equals to loaded tensors in split, throw an exception instead of asserting
* llama_model_loader: ensure mappings vector has the expected size
* llama_model_loader: use at instead of operator[] if this should never add to the map.
* llama_model_loader: immediately add the backend buffer to the model buffers in order to free them if an error occurs in the next allocation. Reserve the expected size.
* llama_model_loader: be sure the model mappings has enough capacity before allocating backend buffer
* llama_model_loader: fix map -> unordered map
* llama_split_prefix: use a clearer version, not pass split path len but dest max len.
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* llama : minor
ggml-ci
* llama : introduce some typedef helpers
* docs: add model shard in hot topic
* llama_model_loader: put mapping in a unique_ptr from the moment it is allocated
Co-authored-by: slaren <slarengh@gmail.com>
* fix llama_split_prefix
---------
Co-authored-by: slaren <slarengh@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* k_cache: be able to use Q5_0
* k_cache: be able to use Q5_1 on CODA
* k_cache: be able to use Q5_0 on Metal
* k_cache: be able to use Q5_1 on Metal
* k_cache: be able to use IQ4_NL - just CUDA for now
* k_cache: be able to use IQ4_NL on Metal
* k_cache: add newly added supported types to llama-bench and CUDA supports_op
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
* server tests : remove seemingly redundant newlines in print()
* server tests : use built-in subprocess features, not os.kill and psutil
* server tests : do not catch e.g. SystemExit; use print_exc
* server tests: handle TimeoutExpired exception
* server tests: fix connect on dual-stack systems
* server: tests: add new tokens regex on windows generated following new repeat penalties default changed in (#6127)
* server: tests: remove the hack on windows since now we get the good socket family
* server: tests: add new tokens regex following new repeat penalties default changed in (#6127)
* server: tests: add new tokens regex following new repeat penalties default changed in (#6127)
---------
Co-authored-by: Pierrick HYMBERT <pierrick.hymbert@gmail.com>
* gguf-split: split and merge gguf files per tensor
* gguf-split: build with make toolchain
* gguf-split: rename `--split-tensors-size` to `--split-max-tensors`. Set general.split_count KV to all split
* split : minor style + fix compile warnings
* gguf-split: remove --upload not implemented
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* backend : offload large batches to GPU
* fix hip
* code cleanup
* fix CUDA split buffers
* Update ggml-backend-impl.h
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* cuda : fix memset without set_device
* imatrix : remove sched affix from weight names
* sched : add a new split if the current one has too many inputs
reduce max inputs per split
more cleanup
* update backends
ggml-ci
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* gritlm: add initial README.md to examples/gritlm
This commit adds a suggestion for an initial README.md for the gritlm
example.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* squash! gritlm: add initial README.md to examples/gritlm
Use the `scripts/hf.sh` script to download the model file.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* squash! gritlm: add initial README.md to examples/gritlm
Fix editorconfig-checker error in examples/gritlm/README.md.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
---------
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
There several places where a gguf context is allocated. A call to gguf_free
is missing in some error paths. Also on linux, llama-bench was missing a
fclose.
* llama : add pipeline parallelism support for batch processing with multiple CUDA GPUs
ggml-ci
* server : add -ub, --ubatch-size parameter
* fix server embedding test
* llama : fix Mamba inference for pipeline parallelism
Tested to work correctly with both `main` and `parallel` examples.
* llama : limit max batch size to n_batch
* add LLAMA_SCHED_MAX_COPIES to configure the number of input copies for pipeline parallelism
default increase to 4 (from 2)
changing this value may improve performance for some systems, but increases memory usage
* fix hip build
* fix sycl build (disable cpy_tensor_async)
* fix hip build
* llama : limit n_batch and n_ubatch to n_ctx during context creation
* llama : fix norm backend
* batched-bench : sync after decode
* swiftui : sync after decode
* ggml : allow ggml_get_rows to use multiple threads if they are available
* check n_ubatch >= n_tokens with non-casual attention
* llama : do not limit n_batch to n_ctx with non-casual attn
* server : construct batch with size of llama_n_batch
* ggml_backend_cpu_graph_compute : fix return value when alloc fails
* llama : better n_batch and n_ubatch comment
* fix merge
* small fix
* reduce default n_batch to 2048
---------
Co-authored-by: Francis Couture-Harpin <git@compilade.net>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* server: format error to json
* server: do not crash on grammar error
* fix api key test case
* revert limit max n_predict
* small fix
* correct coding style
* update completion.js
* launch_slot_with_task
* update docs
* update_slots
* update webui
* update readme
* examples: fix utf8 decoding error
some models have a tokenizer that decodes an id into an incomplete utf8 sequence, need to validate and wait for next token
one example would be: https://huggingface.co/Qwen/Qwen1.5-1.8B-Chat-GGUF/resolve/main/qwen1_5-1_8b-chat-q4_0.gguf and and an example of the token is 18137
* android : minor
---------
Co-authored-by: zhangfuwen <zhangfuwen@foxmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* server: ci: windows build and tests
* server: ci: remove tmp push branch
* server: ci: EOF EOL
* Use builti
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
* server: tests: server graceful shutdown, then kill, then hard kill
* server: tests: remove python2 unicode string
* server: tests: remove wrong comment on server starting, close_fds is always true
* server: tests: server kill, if pid exists
* server: tests: remove dependency to killall
* server: tests: ci windows: pid exists better handling
---------
Co-authored-by: Jared Van Bortel <cebtenzzre@gmail.com>
* add gritlm example
* gritlm results match
* tabs to spaces
* comment out debug printing
* rebase to new embed
* gritlm embeddings are back babeee
* add to gitignore
* allow to toggle embedding mode
* Clean-up GritLM sample code.
* Fix types.
* Flush stdout and output ending newline if streaming.
* mostly style fixes; correct KQ_mask comment
* add causal_attn flag to llama_cparams
* gritml : minor
* llama : minor
---------
Co-authored-by: Douglas Hanley <thesecretaryofwar@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* server: bench: Init a bench scenario with K6
See #5827
* server: bench: EOL EOF
* server: bench: PR feedback and improved k6 script configuration
* server: bench: remove llamacpp_completions_tokens_seconds as it include prompt processing time and it's misleading
server: bench: add max_tokens from SERVER_BENCH_MAX_TOKENS
server: bench: increase truncated rate to 80% before failing
* server: bench: fix doc
* server: bench: change gauge custom metrics to trend
* server: bench: change gauge custom metrics to trend
server: bench: add trend custom metrics for total tokens per second average
* server: bench: doc add an option to debug http request
* server: bench: filter dataset too short and too long sequences
* server: bench: allow to filter out conversation in the dataset based on env variable
* server: bench: fix assistant message sent instead of user message
* server: bench: fix assistant message sent instead of user message
* server : add defrag thold parameter
* server: bench: select prompts based on the current iteration id not randomly to make the bench more reproducible
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* perplexity : support using multiple sequences to allow larger batch sizes
ggml-ci
* set cparams.n_parallel to the number of sequences
* print tested n_ctx, add assert
* add cmake build toggle to enable ssl support in server
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* add flags for ssl key/cert files and use SSLServer if set
All SSL setup is hidden behind CPPHTTPLIB_OPENSSL_SUPPORT in the same
way that the base httlib hides the SSL support
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* Update readme for SSL support in server
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* Add LLAMA_SERVER_SSL variable setup to top-level Makefile
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
---------
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* mamba : begin working on support for Mamba SSM
* mamba : begin figuring out how to (ab)use the kv cache for Mamba
* mamba : recurrent inference almost works, but incoherent
* mamba : recurrent inference WORKS!!!
* convert : optionally use d_conv and d_state from config.json for Mamba
* mamba : refactor recurrent conv, resulting in 20% perf increase
It's still slower than I'd like, but I did not really optimize `ggml_exp` yet.
I also refactored `ggml_exp` to work with tensors with more than 2 dimensions.
* ggml : parallelize ggml_exp
This results in 8% faster token generation for Mamba-130M.
* mamba : simplify the conv step with a self-overlapping view
Turns out the conv_state can be made smaller by one column.
Note that this breaks existing GGUFs of Mamba,
because the key_value_length field is tied to the conv_state size.
Convolution with a self-overlapping view is cool!
And it's much simpler than what I initially thought would be necessary
to make the convolution step work with more than 1 token at a time.
Next step is to make the SSM step work on batches of tokens too,
and thus I need to figure out a way to make a parallel selective scan
which will keep the ssm_state small and won't make it bigger
by a factor of (n_layer * batch_size).
* llama : fix Mamba KV self size wrongly displaying as f16 instead of f32
Relatedly, I also tried to see if other types than f32 worked for the states,
but they don't, because of the operators used.
It's probably better anyway to keep lots of precision there,
since the states are small anyway.
* mamba : fix self-overlapping view depth stride
* mamba : handle batches of more than 1 token
This means running Mamba no longer crashes when using the default settings!
And probably also slightly faster prompt processing.
Both batched and non-batched processing yield the same output.
Previously, the state was not cleared when starting a sequence.
Next step is to make the KV cache API work as expected for Mamba models.
* ggml: add ggml_ssm_scan to help with parallel selective scan
If the selective scan was implemented without a custom operator,
there would be waaay too many nodes in the graph. For example,
for Mamba-130M, with a batch size of 512 (the default),
a naive selective scan could add at least 24*512=12288 nodes,
which is more than LLAMA_MAX_NODES (8192),
and that's only for the smallest Mamba model.
So it's much cleaner with a custom operator.
Not sure about the name, though.
* ggml : in ggml_ssm_scan, merge multiple rows in the same vec operation
This will help with performance on CPU if ggml_vec_mul_f32
and ggml_vec_add_f32 are ever optimized with SIMD.
* mamba : very basic quantization support
Mostly works, but there is currently no difference
between the variants of a k-quant (e.g. Q4_K_S and Q4_K_M are the same).
Most of the SSM-specific weights can be kept in f32 without affecting
the size that much, since they are relatively small.
(the linear projection weights are responsible for most of Mamba's size)
Too much quantization seems to make the state degrade quite fast, and
the model begins to output gibberish.
It seems to affect bigger models to a lesser extent than small models,
but I'm not sure by how much.
Experimentation will be needed to figure out which weights are more important
for the _M (and _L?) variants of k-quants for Mamba.
* convert : fix wrong name for layer norm weight of offical Mamba models
I was using Q-bert/Mamba-* models before, which have a slighlty different
naming scheme for the weights.
(they start with "model.layers" instead of "backbone.layers")
* mamba : fuse more steps of the SSM scan in the ggml_ssm_scan operator
This increases performance on CPU by around 30% for prompt processing,
and by around 20% for text generation.
However, it also makes the ggml_exp and ggml_soft_plus operators unused.
Whether or not they should be kept will be decided later.
* convert : for Mamba, also consider the "MambaLMHeadModel" arch name
It's the name of the class of the official implementation,
though they don't use it (yet) in the "architectures" field of config.json
* mamba : fix vocab size problems with official models
The perplexity was waaaay to high for models with a non-round vocab size.
Not sure why, but it needed to be fixed in the metadata.
Note that this breaks existing GGUF-converted Mamba models,
but **only if** the vocab size was not already rounded.
* ggml : remove ggml_exp and ggml_soft_plus
They did not exist anyway outside of this branch,
and since ggml_ssm_scan fused operations together, they are unused.
It's always possible to bring them back if needed.
* mamba : remove some useless comments
No code change.
* convert : fix flake8 linter errors
* mamba : apply suggestions from code review
* mamba : remove unecessary branch for row-wise ssm_state and C multiplication
It was previously done to avoid permuting when only one token is processed
at a time (like when generating text), but permuting is cheap,
and dynamically changing the compute graph is not future-proof.
* ggml : in ggml_ssm_scan, use more appropriate asserts
* ggml : rename the destination pointer in ggml_compute_forward_ssm_scan_f32
* mamba : multiple sequences, but one at a time
This is a step towards making this Mamba implementation usable
with the server example (the way the system prompt is kept when clearing
the client slots will need to be changed before this can work, though).
The KV cache size for this kind of model is tied to the maximum number
of sequences kept at any single time.
For now, this number is obtained from n_parallel (plus one,
to have an extra sequence to dedicate to the system prompt),
but there might be a better way to do this which won't also
make the main example use 2 cells even if only 1 is really used.
(for this specific case, --parallel 0 helps)
Simultaneous sequence processing will probably require changes to
ggml_ssm_scan, and possibly a new operator for the conv step.
* mamba : support llama_kv_cache_seq_cp
This (mis)uses the logic around K shifts, because tokens in a state
can't be shifted anyway, and because inp_K_shift has the right shape and type.
Using ggml_get_rows is a nice way to do copies, but copy chains can't work.
Fortunately, copy chains don't really seem to be used in the examples.
Each KV cell is dedicated to the sequence ID corresponding to its own index.
* mamba : use a state mask
It's cleaner than the previous heuristic of
checking for the pos of the first token in the batch.
inp_KQ_mask could not be re-used for this, because it has the wrong shape
and because it seems more suited to the next step of
simultaneous sequence processing (helping with the problem of
remembering which token belongs to which sequence(s)/state(s)).
* llama : replace the usage of n_ctx with kv_self.size in many places
* mamba : use n_tokens directly instead of n_tok
* mamba : in comments, properly refer to KV cells instead of slots
* mamba : reduce memory usage of ggml_ssm_scan
From 290.37 MiB to 140.68 MiB of CPU compute buffer size
with Mamba 3B with a batch size of 512.
The result tensor of ggml_ssm_scan was previously a big part
of the CPU compute buffer size. To make it smaller,
it does not contain the intermediate ssm states anymore.
Both y and the last ssm state are combined in the result tensor,
because it seems only a single tensor can be returned by an operator
with the way the graph is built.
* mamba : simultaneous sequence processing
A batch can now contain tokens from multiple sequences.
This is necessary for at least the parallel example, the server example,
and the HellaSwag test in the perplexity example.
However, for this to be useful, uses of llama_kv_cache_seq_rm/cp
will need to be changed to work on whole sequences.
* ggml : add ggml_ssm_conv as a new operator for the conv step of Mamba
This operator makes it possible to use and update the correct states
for each token of the batch in the same way as ggml_ssm_scan.
Other solutions which use existing operators would need loops which would
add too many nodes to the graph (at least the ones I thought of).
Using this operator further reduces the size of the CPU compute buffer
from 140.68 MiB to 103.20 MiB with Mamba 3B with a batch size of 512.
And (at least on CPU), it's a bit faster than before.
Note that "ggml_ssm_conv" is probably not the most appropriate name,
and it could be changed if a better one is found.
* llama : add inp_s_seq as a new input tensor
The most convenient implementation to select the correct state (for Mamba)
for each token is to directly get the correct index from a tensor.
This is why inp_s_seq is storing int32_t and not floats.
The other, less convenient way to select the correct state would be
to have inp_KQ_mask contain 1.0f for each state used by a token
and 0.0f otherwise. This complicates quickly fetching the first used
state of a token, and is also less efficient because a whole row
of the mask would always need to be read for each token.
Using indexes makes it easy to stop searching when there are
no more sequences for a token, and the first sequence assigned
is always very quickly available (it's the first element of each row).
* mamba : support llama_kv_cache_seq_cp copy chains
* mamba : support shifting and dividing the kv cache pos
* mamba : make the server and parallel examples work with whole sequences
A seq_id is dedicated to the system prompt in both cases.
* llama : make llama_kv_cache_seq_rm return whether it succeeded or not
* mamba : dedicate an input tensor for state copy indices
This is cleaner and makes it easier to adapt when/if token positions
(and by extension, inp_K_shift) are no longer integers.
* mamba : adapt perplexity, batched, and batched-bench examples
* perplexity : limit the max number of sequences
This adapts to what the loaded model can provide.
* llama : add llama_n_max_seq to get the upper limit for seq_ids
Used by the perplexity example.
* batched : pass n_parallel to the model's context params
This should have been there already, but it wasn't.
* batched-bench : reserve sequences to support Mamba
* batched-bench : fix tokens being put in wrong sequences
Generation quality isn't what's measured in there anyway,
but at least using the correct sequences avoids using non-consecutive
token positions.
* mamba : stop abusing attention metadata
This breaks existing converted-to-GGUF Mamba models,
but will allow supporting mixed architectures like MambaFormer
without needing to break Mamba models.
This will also allow changing the size of Mamba's states
without having to reconvert models in the future.
(e.g. using something else than d_conv - 1 columns for the conv_states
will not require breaking existing converted Mamba models again)
* gguf-py : add new KV metadata key-value pairs for Mamba
* llama : add new metadata key-value pairs for Mamba
* llama : guard against divisions by zero when n_head is 0
* mamba : rename "unlimited" KV cache property to "recurrent"
* mamba : more correctly update the "used" field of the KV cache
* ggml : in ggml_ssm_scan, use a threshold for soft_plus
This is how the official Mamba implementation does it,
and it's also what torch.nn.Softplus does.
* convert : for Mamba, fallback to internal NeoX tokenizer
The resulting models are exactly the same
as if the tokenizer.json and tokenizer_config.json of GPT-NeoX were there.
* mamba : support state saving and restoring
* ggml : implicitly pass src tensors through dst for Mamba-related ops
* mamba : clarify some comments
* server : fix cache_tokens not getting correctly resized
Otherwise, when the "we have to evaluate at least 1 token" special case
was triggered, an extra token was kept in cache_tokens even if it was
removed from the KV cache.
For Mamba, this caused useless prompt reprocessing when the previous
request triggered the above case.
* convert-hf : support new metadata keys for Mamba
For the models available at
https://huggingface.co/collections/state-spaces/transformers-compatible-mamba-65e7b40ab87e5297e45ae406
* mamba : rename metadata to be more similar to transformers library
This breaks existing converted-to-GGUF models,
but the metadata names are more "standard".
* mamba : support mamba-*-hf models
These models share their token_embd.weight with their output.weight
* mamba : add missing spaces
This is purely a formatting change.
* convert-hf : omit output.weight when identical with token_embd.weight
Only for Mamba for now, but it might be relevant for other models eventually.
Most Mamba models actually share these two tensors, albeit implicitly.
* readme : add Mamba to supported models, and add recent API changes
* mamba : move state_seq and state_mask views outside layer loop
A few tensors were also missing `struct` in front of `ggml_tensor`.
* server : refactoring (wip)
* server : remove llava/clip objects from build
* server : fix empty prompt handling + all slots idle logic
* server : normalize id vars
* server : code style
* server : simplify model chat template validation
* server : code style
* server : minor
* llama : llama_chat_apply_template support null buf
* server : do not process embedding requests when disabled
* server : reorganize structs and enums + naming fixes
* server : merge oai.hpp in utils.hpp
* server : refactor system prompt update at start
* server : disable cached prompts with self-extend
* server : do not process more than n_batch tokens per iter
* server: tests: embeddings use a real embeddings model (#5908)
* server, tests : bump batch to fit 1 embedding prompt
* server: tests: embeddings fix build type Debug is randomly failing (#5911)
* server: tests: embeddings, use different KV Cache size
* server: tests: embeddings, fixed prompt do not exceed n_batch, increase embedding timeout, reduce number of concurrent embeddings
* server: tests: embeddings, no need to wait for server idle as it can timout
* server: refactor: clean up http code (#5912)
* server : avoid n_available var
ggml-ci
* server: refactor: better http codes
* server : simplify json parsing + add comment about t_last
* server : rename server structs
* server : allow to override FQDN in tests
ggml-ci
* server : add comments
---------
Co-authored-by: Pierrick Hymbert <pierrick.hymbert@gmail.com>
* (WIP) Implement stochastic speculative decoding
* sample from residual distribution on draft accept failure
* fix#5657: force greedy sampling with probs when temp is 0
* remove p_accept parameter
* fix style
* remove unused variables
* add srand() in speculative.cpp
* replace use of rand() with mt19937 sampling
* fixes based on review (@JohannesGaessler)
* fix r random generation
* randomly select next sequence to verify + fix bug in memory freeing
* fix bug in active_seqs sync
* fix uniform int distribution initialization
* remove warnings from comparison between int and size_t
* check grammar in `llama_sample_probability_distribution_impl`
* remove malloc code by utilizing vectors
* add PR link to README
* Support special tokens as reverse/anti prompt.
* Tokenize antiprompts only once.
* main : minor
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* server: tests: add models endpoint scenario
* server: /v1/models add some metadata
* server: tests: add debug field in context before scenario
* server: tests: download model from HF, add batch size
* server: tests: add passkey test
* server: tests: add group attention params
* server: do not truncate prompt tokens if self-extend through group attention is enabled
* server: logs: do not truncate log values
* server: tests - passkey - first good working value of nga
* server: tests: fix server timeout
* server: tests: fix passkey, add doc, fix regex content matching, fix timeout
* server: tests: fix regex content matching
* server: tests: schedule slow tests on master
* server: metrics: fix when no prompt processed
* server: tests: self-extend add llama-2-7B and Mixtral-8x7B-v0.1
* server: tests: increase timeout for completion
* server: tests: keep only the PHI-2 test
* server: tests: passkey add a negative test
* suport multiple cards: split-mode - layer|row
* rm warning
* rebase with master, support tow new OPs, close feature for -sm=row, fix for unit test
* update news
* fix merge error
* update according to review comments
* Add "/chat/completions" as alias for "/v1/chat/completions"
* merge to upstream master
* minor : fix trailing whitespace
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Try IQ4_NL with blocks of 64 - does not look good
* iq4_xs: go to super-blocks of 256 and 6-bit scales for blocks of 32
* iq4_xs: CUDA works - 133.2 t/s
* iq4_xs: AVX2 dot product
* iq4_xs: ARM_NEON dot product
* iq4_nl: Metal implementation
As usual, Metal / Apple Silicon don't like my quants.
* iq3_xs: minor fix
* iq4_xs: shrink by using IQ3_S for attn_k and attn_q
* iq4_xs: revert using IQ3_S for attn_k and attn_v
PPL vs size is good, but CPU performance suffers: on M2 Max
TG-128 drops to 21.7 t/s from 28.8, and on a Ryzen-7950X
to 14.5 t/s from 15.8 t/s. On CUDA we have 135 t/s when
using IQ3_S vs 133 t/s with pure IQ4_XS.
* Fix CI
* iq4_xs: Added forgotten check for 256 divisibility
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
