Add near-complete Codex VSCode Support, full OAI Responses bridge#3
Open
michaelw9999 wants to merge 770 commits into
Open
Add near-complete Codex VSCode Support, full OAI Responses bridge#3michaelw9999 wants to merge 770 commits into
michaelw9999 wants to merge 770 commits into
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d72b0819db
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| {"sequence_number", seq_num++}, | ||
| {"output_index", output_idx++}, | ||
| {"item", json { | ||
| {"id", oai_resp_fc_item_id}, |
There was a problem hiding this comment.
server_task_result_cmpl_partial::update() only assigns state.oai_resp_fc_item_id after snapshotting state into the chunk fields, so to_json_oaicompat_resp() can emit response.output_item.added with {"id": oai_resp_fc_item_id} from the previous value (often empty on the first streamed tool call). This makes streamed response.function_call_arguments.delta.item_id/final output_item.done.item.id inconsistent with the announced item, which breaks clients that stitch function-call argument deltas by item_id.
Useful? React with 👍 / 👎.
Comment on lines
+487
to
+489
| if (checkpoints.empty()) { | ||
| return true; | ||
| } |
There was a problem hiding this comment.
When checkpoints is empty, slot_checkpoints_save() returns without touching <filepath>.checkpoints, so reusing the same save filename can leave an old sidecar file behind. A later restore will then load stale checkpoint metadata for a different KV snapshot, which can trigger invalid recurrent-state restore attempts or unnecessary full prompt reprocessing.
Useful? React with 👍 / 👎.
* chore(ui): pin package versions to currently installed - Update all dependencies and devDependencies to match exactly what's in package-lock.json - This ensures reproducible builds by locking to specific versions rather than semver ranges * chore: Update packages * chore: Move remaining dependencies to devDependencies * fix: Add missing `mermaid` package * chore: Update `cookie` package to `v1.1.1` * chore: Formatting * test: Update test configs
…gml-org#22445) * Deduplicate imatrix loading code * Add back LLAMA_TRACE, early exit on quantize missing metadata
…debar (ggml-org#23132) * use child snippets for landing and chat message elements * make ... icon visible in conversation history menu * conversation history forward tab fix * add snippet fix for fork icon in conversation history * focus/keyboard fix for attachment x icon and scroll left/right * formatting * fix scroll down issue * simply Statistics and pointer events in scrolldown * create storybook tests and move to folder * improve tests to actually assert on element
mmvq: Port the ncols_dst optimization from ggml-cuda/mmvq.cu to SYCL. Read weights once per dispatch instead of once per column. Covers all standard quant types + reorder paths for Q4_0, Q8_0, Q3_K, Q4_K, Q5_K, Q6_K. IQ types (except IQ4_XS) excluded due to incompatible vec_dot signatures. ggml-sycl: The weight reorder was only bootstrapped on single-token mat-vec (ne[1] == 1). Speculative / MTP verify issues only multi-column mat-vec, so it never triggered the reorder and ran on the slower non-reorder kernel. Bootstrap it on small multi-column batches (ne[1] <= 8) too.
This PR attempts to slim down the dependencies for build-msys jobs making the same changes that we applied in whisper.cpp to reduce the size of the github actions cache, and should also improve the run time due to fewer dependencies that need to be installed. I realize this is a scheduled job but I think it would still make sense to apply these changes. Refs: ggml-org/whisper.cpp#3858
* Enroll mul_mat_vec_q_moe into PDL, boosting MTP performance on BW
Data collected on a B4500:
Before
```
(llama.cpp) ➜ llama.cpp git:(master) ✗ python mtp-bench.py
code_python pred= 192 draft= 150 acc= 116 rate=0.773 tok/s=202.8
code_cpp pred= 192 draft= 147 acc= 117 rate=0.796 tok/s=212.8
explain_concept pred= 192 draft= 161 acc= 110 rate=0.683 tok/s=196.4
summarize pred= 192 draft= 138 acc= 122 rate=0.884 tok/s=226.6
qa_factual pred= 192 draft= 138 acc= 121 rate=0.877 tok/s=225.1
translation pred= 192 draft= 158 acc= 112 rate=0.709 tok/s=201.5
creative_short pred= 192 draft= 160 acc= 110 rate=0.688 tok/s=197.2
stepwise_math pred= 192 draft= 150 acc= 115 rate=0.767 tok/s=209.2
long_code_review pred= 192 draft= 148 acc= 116 rate=0.784 tok/s=208.9
```
After
```
(llama.cpp) ➜ llama.cpp git:(master) ✗ python mtp-bench.py
code_python pred= 192 draft= 150 acc= 116 rate=0.773 tok/s=211.9
code_cpp pred= 192 draft= 147 acc= 117 rate=0.796 tok/s=224.6
explain_concept pred= 192 draft= 161 acc= 110 rate=0.683 tok/s=207.8
summarize pred= 192 draft= 138 acc= 122 rate=0.884 tok/s=240.2
qa_factual pred= 192 draft= 138 acc= 121 rate=0.877 tok/s=238.5
translation pred= 192 draft= 158 acc= 112 rate=0.709 tok/s=213.4
creative_short pred= 192 draft= 160 acc= 110 rate=0.688 tok/s=208.8
stepwise_math pred= 192 draft= 150 acc= 115 rate=0.767 tok/s=221.7
long_code_review pred= 192 draft= 148 acc= 116 rate=0.784 tok/s=220.7
```
Server launched with:
```
➜ llama.cpp git:(osimons/enroll_mul_mat_vec_q_moe_into_PDL) ✗ ./build-x64-linux-gcc-reldbg/bin/llama-server \
-m /mnt/share/gguf/unsloth/Qwen3.6-35B-A3B-MTP-GGUF/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf -dio \
--spec-type draft-mtp \
--spec-draft-n-max 2 \
-ngl all \
-fa on \
--host 0.0.0.0 \
--port 8080 -np 1 --chat-template-kwargs "{\"preserve_thinking\": true}"
```
* LC to overlap with following kernels
* hparams : refactor hparams.n_layer * cont : remove `n_layer_kv()`, use n_layer_all instead * cont : type consistency * pi : update SYSTEM.md * models : fix Step3.5 MTP * cont : remove duplicate switch cases * cont : explicitly set `false` to extra layers for `is_swa` and `is_recr` * cont : fix nextn layer count handling Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> --------- Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Update quantization readme * install requirements * Apply suggestions from code review Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> * dos2unix suggestions --------- Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
The current link is to a non-existent file. I had a look at the repo, spotted the file containing the UI configuration key and updated the link
* TP: round up granularity to 128 * remove assert
* feat(convert): Get language model conversion working for 4.1 vision Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat(convert): Skip multimodal tensors for GraniteMoeHybrid (vision 4.0) Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Disable vocab padding for non-hybrid models that use GraniteMoeHybrid Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * feat: Add python side architecture name Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: Add python-side plumbing for setting FEATURE_LAYERS hparam Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * feat(mtmd): Add architecture label plumbing Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * feat: Full conversion for mmproj w/ tensor mappings Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Add lm_head skip for mmproj for 4.0 Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: De-alias text_config architecture in convert_lora_to_gguf.py Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: De-alias model.language_model. -> model. for lora adapters Branch: Granite4Vision AI-usage: full (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Extend language model tensor dealiasing in adapters Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Remove unnecessary registration for GraniteSpeech in language model Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: Plumb through mm prefix formatting for qformer tensors Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * feat: Add spatial offests array hparam conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: Add stub plumbing for granite vision in mtmd Branch: Granite4Vision AI-usage: draft (OpenCode + qwen3.5:122b) Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Move deepstack_layer_arr to llm hparam instead of mmproj Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Remove IS_DEEPSTACK_LAYERS This appears to have been added during Qwen3 VL (ggml-org#16780), but it was never actually used. Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: add missing vision attn layernorm eps Branch: Granite4Vision AI-usage: full (OpenCode + Qwen 3.6-35B) Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: Hoist qformer tensors into qf_block and hold a vector for multi-proj Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * feat: Implement c++ hparam parsing Branch: Granite4Vision AI-usage: draft (Claude Code) Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Flatten pinpoints in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Add missing break Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: No reason to have modality prefix for img_pos Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * feat: Add tensor loading Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix(convert): Fix confusion between proj.norm and proj.qformer.layernorm Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Make sure input embeddings are cont before f_embedding_scale Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Invert (h, w) -> (w, h) pinpoints Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> * fix: Fix mmproj hparams in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> * fix: Fix ordering/logic for deepstack injection in granite Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> * fix: Fix preprocessing config to match what the model needs Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> * 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 <ghart@us.ibm.com> Co-authored-by: Eli Schwartz <eliyahu.schwartz@ibm.com> * 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 <ghart@us.ibm.com> * feat: invert embedding multiplier -> base_scale at load Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Fix setting image_resize_pad after new enum introduced Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Add G4V to mmproj mapping in conversion Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Re-add padding disable for non-hybrid hybrid models Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * style: Comment improvement Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: granite_vision -> granite4_vision Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Fix Flake8 warnings in granite conversion module Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Fix GGUF key for deepstack_layers_arr Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * refactor: deepstack_layers(_arr) -> deepstack_mapping(_arr) Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Remove unnecessary ggml_cont and build_forward_expand in cbx Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * style: Clean up comments Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Remove unnecessary `unordered_set` include Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Add architecture guard on deepstack_mapping_arr printout Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Remove unnecessary AI-gen comment Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * style: Remove TODO about block/vs non-block tensor mapping Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: Move is_vision_feature_layer logic into clip_hparams Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: Use a bool for append_token Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * style: Remove unnecessary comment Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Remove unused get_model api yikes! Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> * fix: Fix norm/post_norm mixup in conversion face. palm. :( Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * style: More descriptive tensor names Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix: Apply PR cleanup for new conversion changes AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> * fix(convert): Remove duplicate V_ENC_EMBD_IMGNL Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * refactor: append_token -> add_newline Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * style: Comment cleanup Branch: Granite4Vision AI-usage: none Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * 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 <ghart@us.ibm.com> --------- Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* vulkan: add fwht support for Intel with shmem reduction * don't use N as workgroup size * disable subgroup shuffle on MoltenVK AMD * disable fwht shader on Intel Windows due to driver bug
…put_tokens API (ggml-org#23913) * mtmd: add "placeholder bitmap" for counting tokens w/o preprocessing * fast path skip preproc for placeholder * fix build * correct the api * add server endpoint + tests * add object name * update docs * add proxy handling * fix build * fix audio input path * use is_placeholder in process_mtmd_prompt() * nits * nits (2) * docs: clarify chat/completions/input_tokens is not official * fix merge problem
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
Flatten the partition over n_batch * M so every thread participates in
the quantization
| CPU | Model | Test | t/s OLD | t/s NEW | Speedup |
|:--------------------------------|:------------------------------|:-------|----------:|----------:|----------:|
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B IQ4_NL - 4.5 bpw | pp512 | 730.71 | 779.86 | 1.07 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B IQ4_NL - 4.5 bpw | tg128 | 87.88 | 86.79 | 0.99 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B IQ4_XS - 4.25 bpw | pp512 | 725.09 | 1023.31 | 1.41 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B IQ4_XS - 4.25 bpw | tg128 | 83.64 | 83.62 | 1.00 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_0 | pp512 | 820.51 | 924.05 | 1.13 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_0 | tg128 | 90.59 | 92.46 | 1.02 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_1 | pp512 | 776.88 | 872.79 | 1.12 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_1 | tg128 | 89.39 | 90.94 | 1.02 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_K_M | pp512 | 719.28 | 1009.27 | 1.40 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_K_M | tg128 | 80.62 | 80.86 | 1.00 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_K_S | pp512 | 732.29 | 1077.29 | 1.47 |
| Intel(R) Xeon(R) Platinum 8488C | qwen35 0.8B Q4_K_S | tg128 | 86.42 | 83.53 | 0.97 |
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
GLM-5.2 ships the DSA "lightning indexer" on only a subset of layers (the "full" layers; others omit it), but the GLM_DSA loader created the five indexer tensors on every layer as required, so loading any GLM-5.2 GGUF failed with e.g. `missing tensor 'blk.3.indexer.k_norm.weight'`. GLM_DSA's graph is llama_model_deepseek2::graph (plain MLA) and does not use the indexer tensors (indexer runtime not yet implemented), so they are loaded-but-unused. Marking them TENSOR_NOT_REQUIRED lets layers without an indexer load as nullptr and the model runs as full MLA attention. DeepSeek-V3.2 (uniform indexer on all layers) is unaffected.
* server: avoid forwarding auth headers in CORS proxy * format * fix test * fix e2e test --------- Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* arg: try fixing test-args-parser randomly fails * return ref * try triggering the workflow * exception wrapper * wip * test * test 2 * arg: guard win32 utf8 argv override make_utf8_argv rebuilds argv from GetCommandLineW to fix utf8 handling of non ascii arguments on windows. the override runs unconditionally inside common_params_parse, so it also clobbers a programmatic argv passed by a caller. test-arg-parser builds a synthetic argv but then sees the real process command line instead, the model argument is never parsed, and the assert that expects success aborts via fastfail (0xC0000409). this shows up as a random failure in the openvino windows workflow. only override argv when its length matches the caller argc, so the utf8 repair still applies to real binaries while a programmatic argv stays intact. --------- Co-authored-by: Pascal <admin@serveurperso.com>
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
* common/peg : refactor until gbnf grammar into an ac automaton * cont : add a test with multiple strings * cont : pad state with 0s so rules line up * cont : clean up comments * cont : use set everywhere * cont : inline state num string padding * cont : add a ref to PR * cont : fix regression in server-tools.cpp
* add mtp_layer_offset + include nextn flags in graph reuse * add llama_set_mtp_layer_offset + llama_model_n_nextn_layer API * offset head select + require all MTP blocks * speculative multi-head process() * speculative multi-head draft() * gather outputs via inp_out_ids * cleanup * fix core * minor cleanup * merged draft_multi_head into draft() * mtp rename nextn * Apply suggestions from code review Co-authored-by: Aman Gupta <amangupta052@gmail.com> * clean-up comments * fix for multi seq * apply suggestions && chain-heads comment * add a reference for chain_heads discussion --------- Co-authored-by: Aman Gupta <amangupta052@gmail.com>
…org#24828) * server: real-time model load progress tracking via /models/sse * update docs * add mutex for notify_to_router * correct docs
* implement call statement * undo unintended change * de-lambda * simplify * move caller context inside function handler
* server: refactor batch construction * wip * wip 2 * wip 3 * wip 4 * add abort_all_slots * handle batch full more carefully * fix assert * rm debug log * small nits * (debug) add timings * debug: force llama_synchronize for accurate timings * address comments * disable DEBUG_TIMINGS
ggml-org#24870) * server: fix report progress for loading spec models, add "stages" list * improve * nits * nits 2
…l-org#24869) * common/peg : implement ac parser * cont : extract functions * cont : tidy up * cont : remove a test * cont : move ac() def
…gml-org#24893) line_start -1 normalized to n+1, so append inserted at lines.begin() + n + 1, one past end() -> heap-buffer-overflow in vector::_M_range_insert. Normalize -1 to n (insert at end()), restrict -1 to append mode and reject it for replace/delete instead of silently clobbering the last line. Parenthesize the insert offset so empty-file append computes the position as int first, avoiding a transient begin() - 1 on a null vector data pointer.
* support bf16 on bin_bcast OP and unary OPs * support the older Intel compiler than 2026.0
* ui: model status and load progress via /models/sse feed * ui: centralize SSE wire-format delimiters into shared constants for the chat and /models/sse parsers * ui: type /models/sse event names as a ServerModelsSseEventType enum Address review from allozaur
* server: refactor/generalize input file schema * wire up input_video, accept raw base64 * nits * nits (2) * fix windows
…g#24834) * server: real-time model load progress tracking via /models/sse * update docs * server: move model download to child process * rm unused * fix most problems * clean up * nit fixes * fix test case * do not detact() thread * shorter MODEL_DOWNLOAD_TIMEOUT in test * throttle
Updated model selection prioritization to include favorite models.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Things brings in automatic compaction,
web_searchandfile_searchand is super easy to configure, for example:For the automatic compaction to work, you must set
model_context_windowandmodel_auto_compact_token_limit. Summary boxes and clickable diffs with the undo button ususally needmodel_supports_reasoning_summaries = trueandmodel_reasoning_summary = "auto".Just install tavily (but shell command is
tvly) andrgor any other preferred web search MCP or file search/locator tool, it will wrap it through the shell and integrate it more natively and intuitively. If left out, it will hide these tools from the model.