* bitnet: put token embeddings on the GPU
* Update README with the new CUDA/Meat performance
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Also save one scale operation in the ffn network by adjusting
rms_eps. We gain up to 3% in performance by doing this, but it
is a bit of a hack (we store the tensor scales in op_params
while loading the model).
This reverts commit f83381371b61e0863b55c60e5f5df139126a496d.
When using CUDA, the tensor contents have not been loaded yet,
so we crash when trying to access the scale when building the
graph. There must be a better way.
and correspondingly add an extra ggml_mul_mat operation.
As per @ggerganov, this is how things should be done.
It seems to be working, but as far as I can tell this
results in a ~15% performance penalty for prompt processing.
Commiting so I can go and test on othe platforms.
* create append_pooling operation; allow to specify attention_type; add last token pooling; update examples
* find result_norm/result_embd tensors properly; update output allocation logic
* only use embd output for pooling_type NONE
* get rid of old causal_attn accessor
* take out attention_type; add in llama_set_embeddings
* bypass logits when doing non-NONE pooling
* update: convert-hf-to-gguf.py to support Qwen2-57B-A14B
* fix: QWEN2MOE support for expert_feed_forward_length
previously, expert ff was taken from n_ff (intermediate size) but it is now properly taken from LLM_KV_EXPERT_FEED_FORWARD_LENGTH
n_ff_exp and n_ff_shared_exp are now properly calculated
* update: convert-hf-to-gguf.py cleanup for Qwen2MoeForCausalLM
* fix: QWEN2MOE support for expert_feed_forward_length
previously, expert ff was taken from n_ff (intermediate size) but it is now properly taken from LLM_KV_EXPERT_FEED_FORWARD_LENGTH
n_ff_exp and n_ff_shexp are now properly calculated
* Implement non-mapped async IO for CUDA on Windows. On a fast Gen5 NVMe drive this change improves model load time by >3x while it should be the same (or slightly faster) on any other drive.
* Free resources except for backend.
* Change assertions to exceptions in llama_file, find correct cuda backend to create CUDA resources and respect the use_mmap flag again for CUDA.
* Apply suggestions from code review
Co-authored-by: slaren <slarengh@gmail.com>
* Fix editorconfig and unused variable
* Fix issues with Windows build
---------
Co-authored-by: slaren <slarengh@gmail.com>
* separate DPCT helpers outside
* replace global variables with context
* remove useless extra
* update mul_mat condition
* remove duplicate buft initialization
* remove duplicate extra and global work group size
* remove useless backend check
* remove duplicated extras
* use macro for group_size and remove cuda-related
* support for Poro chat pre-tokenizer
* add support for Poro pre-tokenizer
* Update convert-hf-to-gguf-update.py
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Change Poro-34B-chat to poro-chat
* Change Poro-34B-chat to poro-chat
* Update convert-hf-to-gguf-update.py
* Update llama.cpp
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* move BLAS to a separate backend
* rename GGML_USE_OPENBLAS to GGML_USE_BLAS
* alloc : reuse same buffer when the same buffer type if used multiple times
* set number of threads automatically for openblas and blis
* sched : print assignments when GGML_SCHED_DEBUG env variable is set
* sched : allow ops with weights on an incompatible buffer type
This will cause the weight to be copied to a backend that supports the
op, which is very costly. The weight should have been stored in a buffer
of a backend that can run the op, but llama.cpp cannot do this
automatically at the moment.
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* feat: add changes to handle jina v2 base code
* fix: do not complicate things
* fix: fix the usage of the code model
* fix: fix comments
* fix: fix linting issues
* fix: remove ollama patches
* style : minor
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* common : gpt_params_parse do not print usage
* common : rework usage print (wip)
* common : valign
* common : rework print_usage
* infill : remove cfg support
* common : reorder args
* server : deduplicate parameters
ggml-ci
* common : add missing header
ggml-ci
* common : remote --random-prompt usages
ggml-ci
* examples : migrate to gpt_params
ggml-ci
* batched-bench : migrate to gpt_params
* retrieval : migrate to gpt_params
* common : change defaults for escape and n_ctx
* common : remove chatml and instruct params
ggml-ci
* common : passkey use gpt_params
* Add per token attributes enum
* Using phi-3 for testing 'rstrip'
* Using jina-v2 for testing 'lstrip'
* Brute force test for 'lstrip' and 'rstrip'
* Implement 'rstrip' and 'lstrip'
* Update phi-3 GGUF file (obsolete since 917dc8c)
* Replace llama_token_type with llama_token_attribs
* llama : offload to RPC in addition to other backends
* - fix copy_tensor being called on the src buffer instead of the dst buffer
- always initialize views in the view_src buffer
- add RPC backend to Makefile build
- add endpoint to all RPC object names
* add rpc-server to Makefile
* Update llama.cpp
Co-authored-by: slaren <slarengh@gmail.com>
---------
Co-authored-by: slaren <slarengh@gmail.com>
* llama : cache llama_token_to_piece
ggml-ci
* llama : use vectors and avoid has_cache
ggml-ci
* llama : throw on unknown tokenizer types
ggml-ci
* llama : print a log of the total cache size
* Update random test: add_bos_token.
* Update random test: add WPM models for testing.
* Build vocab.special_tokens_cache using vocab token types.
* Fix and improve WPM preprocessing.
- Fix unicode edge case combinations.
- Split by whitspace in the same pass.
* Discard all tokens when no matching found.
* Add optional MLP bias for Granite models
Add optional MLP bias for ARCH_LLAMA to support Granite models.
Partially addresses ggerganov/llama.cpp/issues/7116
Still needs some more changes to properly support Granite.
* llama: honor add_space_prefix from the model configuration
propagate the add_space_prefix configuration from the HF model
configuration to the gguf file and honor it with the gpt2 tokenizer.
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* llama: add support for small granite models
it works only for the small models 3b and 8b.
The convert-hf-to-gguf.py script uses the vocabulary size of the
granite models to detect granite and set the correct configuration.
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
---------
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
Co-authored-by: Steffen Roecker <sroecker@redhat.com>
* common : increase max number of experts to 160
* common : add tensors ATTN_Q_A, ATTN_Q_A_NORM, ATTN_Q_B, ATTN_KV_A_MQA, ATTN_KV_A_NORM, ATTN_KV_B needed by DeepSeek-V2 MLA (multi-head latent attention) architecture
* common : add model header parameters: leading_dense_block_count, expert_feed_forward_length, expert_shared_count, expert_weights_scale, attention.q_lora_rank, attention.kv_lora_rank, rope.scaling.yarn_log_multiplier
* convert-hf : add model conversion support for DeepseekV2ForCausalLM
* llama : add model types for DeepSeek-V2 and DeepSeek-V2-Lite models
* llama : add two new llm_build_moe_ffn() arguments: scale_w (whether to scale weights of selected MoE experts) and w_scale (numerical value of the scaling factor)
* llama : add inference support for LLM_ARCH_DEEPSEEK2
---------
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
* main : don't print special tokens with --grammar
The CLI interface was recently changed to print special control tokens
like the </s> stop message one. This token shouldn't be printed if the
grammar flag was passed, unless the grammar specifies it, because that
breaks shell-scriptability.
* main: use seperate stream for control characters
* main: use dprintf and add --ctrl-token-no-out and --ctrl-token-fd-out
* main: dprintf isn't part of the IEEE POSIX standard. Just use write().
* main: remove --ctrl-token-fd-out in favor for fcntl() based detection
* common.cpp: accidentally removed --interactive-first
* main: only merge stdout and control token if not in conversation or grammar mode
* main: rejig control token descriptor handling
* main: must check pipe status on very top of program
* main: renamed --no-special from --ctrl-token-no-out and other refactoring
* main: refactor ctrl_token_no_out --> no_special
* llama: rename llama_token_is_control_token() to llama_token_is_control()
* main: remove special token file descriptor feature (#5)
---------
Co-authored-by: Brian <mofosyne@gmail.com>
* common : increase max number of experts to 128
* common : add tensor LLM_TENSOR_FFN_NORM_EXPS for normalization before MoE that runs in parallel to attention + ffn
* gguf-py : add architecture-specific block mappings that override selected general block mappings
* convert-hf : add model conversion support for ArcticForCausalLM
* convert-hf : use added_tokens_decoder from tokenizer_config.json to redefine tokens from SentencePiece model (only for ArcticForCausalLM)
* llama : add inference support for LLM_ARCH_ARCTIC
---------
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
* Fix phi3 template matching vs zephyr
* Add regression test for new phi3 chat template
* Implement review suggestions
* Fix phi3 jinja test templates & match by <|end|>
* Apply suggestion
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* Add all phi3 template variants in tests
* Remove unneeded message trimming
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* Fix tests to not expect trimmed messages
---------
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
