mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-09-04 02:37:27 +02:00
Gabe Goodhart
64086f2b2f
* feat(convert): Get language model conversion working for 4.1 vision Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat(convert): Skip multimodal tensors for GraniteMoeHybrid (vision 4.0) Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Disable vocab padding for non-hybrid models that use GraniteMoeHybrid Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Plumb python-side vision projector names and mappings There are several awkward things here: 1. Most of these are essentially identical to the audio qformer tensors. On the c++ side, that's mapped using the prefix, so the rest of the GGUF name needs to align, but on the python side there's no prefix notion, so they all get duplicated. 2. There are a couple of net-new tensors for vision, in particular PROJ_NORM. In both speech and vision, the QF_PROJ_NORM is qualified as belonging to the qformer portion, but the GGUF name is simply proj_norm which conflicts with the ideal name for this new PROJ_NORM that is not qualified as part of the qformer. To get around this, I used "proj_layernorm" as the GGUF name. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add python side architecture name Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add python-side plumbing for setting FEATURE_LAYERS hparam Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add c++ side tensor naming defines NOTE: Usage of these hasn't been updated to include prefix yet Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat(mtmd): Convert vision_feature_layer to an ordered vector We need to preserve the ordering of these feature index values so that they can be mapped to the sub-tensors within the stacked projectors. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat(mtmd): Add architecture label plumbing Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <[email protected]> * feat(wip): Add partial conversion for mmproj This handles stacking the projector tensors and setting the new harams Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add gguf_writer and constant support for new hparams and deepstack layer arr Branch: Granite4Vision AI-usage: draft (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <[email protected]> * feat: Full conversion for mmproj w/ tensor mappings Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add lm_head skip for mmproj for 4.0 Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: De-alias text_config architecture in convert_lora_to_gguf.py Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add --trust-remote-code arg to convert_lora_to_gguf.py This defaults to False, but allows a user to enable it programmaticly instead of using the interactive prompt. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: De-alias model.language_model. -> model. for lora adapters Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <[email protected]> * fix: Extend language model tensor dealiasing in adapters Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove unnecessary registration for GraniteSpeech in language model Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Plumb through mm prefix formatting for qformer tensors Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Refactor vision projector tensors to use predictor ID as the block This is cleaner than stacking them. The modeling file hard-codes single-layer qformers, so we can punt on the multiipule multi-layer projectors problem. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add spatial offests array hparam conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add stub plumbing for granite vision in mtmd Branch: Granite4Vision AI-usage: draft (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add new hparam and tensor naming in clip-impl.h New hparams: - KEY_PROJ_SAMPLE_QUERY_SIDE - KEY_PROJ_SAMPLE_WINDOW_SIDE - KEY_PROJ_SPATIAL_OFFSETS New tensors: - TN_MULTI_PROJ_IMG_POS - TN_MULTI_PROJ_QUERY - TN_MULTI_PROJ_LAYERNORM - TN_MULTI_PROJ_LINEAR - TN_MULTI_PROJ_NORM Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Move deepstack_layer_arr to llm hparam instead of mmproj Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove IS_DEEPSTACK_LAYERS This appears to have been added during Qwen3 VL (https://github.com/ggml-org/llama.cpp/pull/16780), but it was never actually used. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: n_deepstack_layers -> deepstack_layer_arr The old logic hard coded a correspondence between the first N layers of the LLM and the 1->N entries in the input embeddings. Now, that relationship is maintained at loading time if the GGUF value is single-valued. If it is multi-valued, it loads directly allowing for deepstack layers to be spaced out throughout the model. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Use try/catch for single/multi valued deepstack info The alternative would be to use get_key_or_arr, but then the single value would be populated through the entire array and we'd need to detect that and update it with the right correspondence. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add deepstack injection point for granite LLM The use of ggml_add here assumes that the elements of inp_embd will be pre- arranged to be the full embedding length with only the vision-mask'ed portions non-zero from the projector. This matches how Qwen3VL does it. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: add missing vision attn layernorm eps Branch: Granite4Vision AI-usage: full (OpenCode + Qwen 3.6-35B) Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Hoist qformer tensors into qf_block and hold a vector for multi-proj Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix missing prefix template for TN_QF_PROJ_LINEAR It's not strictly necessary since vision uses the blockwise version, but it makes the loading consistent. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add embedding scale and image grid pinpoints hparams in conversion Also remove dead parsing for self._deepstack_layer_arr Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add mtmd KEY_ section for hparams shared with the LLM In this case, we need the EMBEDDING_SCALE so we can unscale the image embeddings to compensate for applying embedding scale to the input embeddings Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Implement c++ hparam parsing Branch: Granite4Vision AI-usage: draft (Claude Code) Co-authored-by: Eli Schwartz <[email protected]> Signed-off-by: Gabe Goodhart <[email protected]> * fix: Flatten pinpoints in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add missing break Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: No reason to have modality prefix for img_pos Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add tensor loading Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix(convert): Fix confusion between proj.norm and proj.qformer.layernorm Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Use the right portion of speech for tensor loading! Also plumb through the layernorm -> post_norm naming change Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add logging of deepstack_layers_arr if set I also changed the print_f output type to int32_t to avoid printing overflow values for -1. This could cause overflows on the other side, but I can't imagine a value for any of the current array hparams that would trigger that. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Make sure input embeddings are cont before f_embedding_scale Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add init and mmproj_embd cases for g4v The n_mmproj_embd is 1+ to make space for the text embedding and all 8 projectors Branch: Granite4Vision AI-usage: draft (Bob) Signed-off-by: Gabe Goodhart <[email protected]> * fix: Invert (h, w) -> (w, h) pinpoints Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Reorder projectors based on llm index and skip the first injection The multi-projector stack has a strange asymmetry based on how it's currently implemented for qwen3vl: on the mmproj side, it's all N projectors, but the output of the "first" (by inp_embd index) projector is automatically consumed as if it were a standard single-projector mmproj, so the deepstack portion needs to only contain the 1-N entries. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Eli Schwartz <[email protected]> * fix: Fix mmproj hparams in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Eli Schwartz <[email protected]> * fix: Fix ordering/logic for deepstack injection in granite Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Eli Schwartz <[email protected]> * fix: Fix preprocessing config to match what the model needs Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Eli Schwartz <[email protected]> * wip: Partial port of Eli's implementation This is still pretty broken, but it's getting closer. It now happily generates tokens, but the values are quite incorrect still. I suspect it's caused by the mapping of projectors from safetensors to their respective orders here. Also, this implementation breaks encapsulation pretty badly in mtmd_encode. This will need a big refactor to put the G4V-specific encoding logic somewhere more appropriate. Branch: Granite4Vision AI-usage: draft (Claude Code, Bob) Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Eli Schwartz <[email protected]> * fix: Fix the pre-scaling on the input embeddings to correctly invert the scale We've got tokens! They still don't line up quite right, so something's a little off, but we're getting much closer now. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: invert embedding multiplier -> base_scale at load Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix setting image_resize_pad after new enum introduced Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add G4V to mmproj mapping in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Re-add padding disable for non-hybrid hybrid models Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Simplify G4V n_tokens computation This is slightly more efficient and flexible for when we implement the unpad cropping. IMO, it's also clearer that it is adding the number of image_newline tokens (embeddings) to the grid, rather than recomputing the entire count. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add new clip APIs for post-tile-encoding assembly Granite 4 Vision uses llava-next style pack-and-unpad which requires injecting the learned newline after each row of the tile grid. A row here is a single row of the grid which is composed of (grid_x * cols_per_tile) * (grid_y * rows_per_tile), so the result is newlines injected in between individual tile rows, thus not something that can be handled with the standard llava-uhd block-wise endcoding. Branch: Granite4Vision AI-usage: draft (Claude Code + Opus 4.7) Signed-off-by: Gabe Goodhart <[email protected]> * feat: Add model interfaces for granite 4 vision assembler I'm on the fence about the best organization of this. These free functions allow the per-architecture logic in clip.cpp to access the model-specific graph building, but they still require a fair bit of model-specific logic in clip.cpp which is not ideal. I think a better approach may be to replicate what is done with the graph builders themselves (and possibly even make the assembler part of the model's existing graph builder). Branch: Granite4Vision AI-usage: full (Claude Code + Opus 4.7) Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Remove all g4v-specific branching from mtmd.cpp in favor of clip assembler Branch: Granite4Vision AI-usage: full (Claude Code + Opus 4.7) Signed-off-by: Gabe Goodhart <[email protected]> * refactor(mtmd): Consolidate assembler logic into clip_assembler class family Just like `clip_graph` is the base class for building the model-specific encoder graphs, `clip_assembler` will be the base class for building the model-specific assembler graphs. This allows the assembly pattern to follow how the encoder pattern is implemented where the model-specific logic lives in a subclass co-located with the encoder graph builder that gets constructed by a simple factory method. Branch: Granite4Vision AI-usage: full (Claude Code + Opus 4.7) Signed-off-by: Gabe Goodhart <[email protected]> * style: Comment improvement Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: granite_vision -> granite4_vision Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove dead codepath for Qwen3VL add_vision_is_deepstack These pieces were never used on the c++ side (removed there in an earlier commit), so this is just cleanup that I missed before. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Oops! I did not mean to commit one of my prompt files But now it's too far back in history to effectively rebase out, even with interactive and --rebase-merges :( Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add missing <algorithm> include for std::find It seems that this was already pulled in on some platforms, but not on others Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix Flake8 warnings in granite conversion module Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Remove clip_assembler in favor of clip_image_f32.append_token Per conversation in the PR, the clip_assembler pattern was too invasive. This is a compromise that limits model-specific blocks to add_media where each preprocessed tile is annotated with an injection type, after which all the token counting logic is generic and the newline injection itself is handled in the graph based on the value for the given tile image. Branch: Granite4Vision AI-usage: draft (Bob, OpenCode + Qwen 3.6 35b) Signed-off-by: Gabe Goodhart <[email protected]> * refactor(convert): Split n_deepstack_layers and deepstack_layers (array) Branch: Granite4Vision AI-usage: full (Bob, OpenCode + Qwen3.6-35b) Signed-off-by: Gabe Goodhart <[email protected]> * refactor(src): Handle n_deepstack_layers and deepstack_layers GGUF keys Branch: Granite4Vision AI-usage: draft (Bob, OpenCode + Qwen3.6-35b) Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix GGUF key for deepstack_layers_arr Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Remove pre-scaling embeddings and skip scaling for raw embd inputs This follows how gemma3 and gemma4 handle embedding scaling by skipping the multiplier for raw input embeddings. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: deepstack_layers(_arr) -> deepstack_mapping(_arr) Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Fully revert changes to n_deepstack_layers and qwen3vl* Since we're going to keep the GGUF KVs separate, it makes sense to just keep the hparams separate too to limit the scope of this branch. The down side is that n_deepstack_layers and deepstack_mapping_arr are potentially conflicting. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Revert removal of "is_deepstack_layers" GGUF KV This KV is not used at all on the c++ side, so it's fully dead, but there's also no need to conflate this cleanup with the addition of G4V. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove unnecessary ggml_cont and build_forward_expand in cbx Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * style: Clean up comments Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Tighter and more flexible code for g4v_build_block This could be refactored to look a lot more like granite-speech, but the overall block constructs before/after the qformer are pretty different, so for now I'm going to leave it as is and just tighten a bit. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove unnecessary `unordered_set` include Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Add architecture guard on deepstack_mapping_arr printout Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove unnecessary AI-gen comment Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Always initialize deepstack_mapping_arr with -1 values This was causing `test-llama-archs` to fail, likely due to trying to save the uninitialized values, then re-loading them. It's safer to always initialize so that other models don't forget and end up with undefined behavior. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * style: Remove TODO about block/vs non-block tensor mapping Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Move is_vision_feature_layer logic into clip_hparams Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Use a bool for append_token Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * style: Remove unnecessary comment Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Remove unused get_model api yikes! Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: Rearrange helpers for g4v to be private members and use build_attn Branch: Granite4Vision AI-usage: full (Bob, OpenCode + Qwen3.6-35b) Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix off-by-one in vision layer index This was inherited from the Claude Code implementation that pushed the negative index inversion down into the model file. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Fix norm/post_norm mixup in conversion face. palm. :( Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * style: More descriptive tensor names Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * fix: Apply PR cleanup for new conversion changes AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> Co-authored-by: Sigbjørn Skjæret <[email protected]> * fix(convert): Remove duplicate V_ENC_EMBD_IMGNL Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * refactor: append_token -> add_newline Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * style: Comment cleanup Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> * feat: Cleaner error handling/checking NOTE: format_string is not available in granite.cpp (and including clip-impl.h to get it doesn't compile, so I think it violates the intended encapsulation), so std::stringstream is the simplest answer. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <[email protected]> --------- Signed-off-by: Gabe Goodhart <[email protected]>