repo stringclasses 15
values | fix_commit stringlengths 40 40 | buggy_commit stringlengths 40 40 | message stringlengths 3 64.3k | files listlengths 1 300 | timestamp timestamp[s]date 2013-03-13 20:45:00 2026-04-11 07:48:46 |
|---|---|---|---|---|---|
ggml-org/llama.cpp | c959b676be29e93f8dbc3bd6056ceba812a9eb72 | cd08fc3ecc0264b4414b68af3874a6c689ed60c1 | CUDA: fix FA occupancy, optimize tile kernel (#15982) | [
{
"path": "ggml/src/ggml-cuda/common.cuh",
"patch": "@@ -75,6 +75,8 @@\n #define GGML_CUDA_CC_IS_RDNA4(cc) (cc >= GGML_CUDA_CC_RDNA4)\n #define GGML_CUDA_CC_IS_GCN(cc) (cc > GGML_CUDA_CC_OFFSET_AMD && cc < GGML_CUDA_CC_CDNA1)\n #define GGML_CUDA_CC_IS_CDNA(cc) (cc >= GGML_CUDA_CC_CDNA1 && cc < GGML_CUDA_... | 2025-09-17T13:32:42 |
vuejs/vue | e4fd6327906a8470c7d9999b0729c24e46cc6b67 | c1bf20acb675aff7ba3e13381b1d45c395a3da7f | avoid resetting innerHTML/textContent if value is the same (fix #4247) | [
{
"path": "src/platforms/web/runtime/modules/dom-props.js",
"patch": "@@ -21,13 +21,14 @@ function updateDOMProps (oldVnode: VNodeWithData, vnode: VNodeWithData) {\n }\n }\n for (key in props) {\n+ cur = props[key]\n // ignore children if the node has textContent or innerHTML,\n // as the... | 2016-11-18T16:56:56 |
huggingface/transformers | de3ee737cf0e47f96c4723b919e920f0b291bd30 | b914445f77a94956d1d6543686a329ec2708cc35 | Fix README.md error when installing from source (#41303) | [
{
"path": "README.md",
"patch": "@@ -111,10 +111,10 @@ git clone https://github.com/huggingface/transformers.git\n cd transformers\n \n # pip\n-pip install .[torch]\n+pip install '.[torch]'\n \n # uv\n-uv pip install .[torch]\n+uv pip install '.[torch]'\n ```\n \n ## Quickstart",
"additions": 2,
"de... | 2025-10-02T23:08:27 |
denoland/deno | f5cb510ebac83583ca95d516d6c689e79c531f51 | dd80d7575f7c07e2eaa73ad76ce3d811fb61f1b0 | fix(bundle): support android (#31521)
support android os in esbuild
linked with https://github.com/denoland/esbuild_client/pull/16 | [
{
"path": "cli/tools/bundle/esbuild.rs",
"patch": "@@ -28,6 +28,8 @@ fn esbuild_platform() -> &'static str {\n (\"aarch64\", \"macos\" | \"apple\") => \"darwin-arm64\",\n (\"x86_64\", \"windows\") => \"win32-x64\",\n (\"aarch64\", \"windows\") => \"win32-arm64\",\n+ (\"x86_64\", \"android\") ... | 2025-12-09T22:28:31 |
ollama/ollama | abfc73d31ed41e0f858199d0d3f6cff882c1c81d | 4cb42ca55e718777c16b2c66340b0d5c3405cca6 | make response errors unique for error trace | [
{
"path": "server/images.go",
"patch": "@@ -510,7 +510,7 @@ func CopyModel(src, dest string) error {\n \t\treturn err\n \t}\n \n-\terr = ioutil.WriteFile(destPath, input, 0644)\n+\terr = ioutil.WriteFile(destPath, input, 0o644)\n \tif err != nil {\n \t\tfmt.Println(\"Error reading file:\", err)\n \t\treturn... | 2023-07-24T19:04:21 |
ggml-org/llama.cpp | cd08fc3ecc0264b4414b68af3874a6c689ed60c1 | cb5bb6cc05119c24e7711ca2956cd0e6d409d396 | common : Fix corrupted memory error on json grammar initialization (#16038)
Initalizing RESERVED_NAME in is_reserved_name() is not thread
safe and leads to corrupted memory when used from multiple threads
as can be seen in the asan trace below. This fixes the initialization
to make it thread-safe.
#0 0x000100abd0... | [
{
"path": "common/json-schema-to-grammar.cpp",
"patch": "@@ -257,12 +257,13 @@ std::unordered_map<std::string, BuiltinRule> STRING_FORMAT_RULES = {\n };\n \n static bool is_reserved_name(const std::string & name) {\n- static std::unordered_set<std::string> RESERVED_NAMES;\n- if (RESERVED_NAMES.empty()... | 2025-09-17T08:08:02 |
vuejs/vue | c1bf20acb675aff7ba3e13381b1d45c395a3da7f | 9a742cb423690acb29990bd38f5c17f7b21f12f8 | do not trigger change on NaN->NaN set (fix #4236) | [
{
"path": "src/core/observer/index.js",
"patch": "@@ -162,7 +162,7 @@ export function defineReactive (\n },\n set: function reactiveSetter (newVal) {\n const value = getter ? getter.call(obj) : val\n- if (newVal === value) {\n+ if (newVal === value || (newVal !== newVal && value !== ... | 2016-11-18T16:47:02 |
huggingface/transformers | 41e5abac5cb49983a08ddef3e8645d6efd23c8f3 | da3c7d1d36a7f4df7c6227ffc82d2bb5d3fa4a57 | FIX: Bug in PEFT integration delete_adapter method (#41252)
The main content of this PR is to fix a bug in the delete_adapter method
of the PeftAdapterMixin. Previously, it did not take into account
auxiliary modules from PEFT, e.g. those added by modules_to_save. This
PR fixes this oversight.
Note that the PR uses a... | [
{
"path": "src/transformers/integrations/peft.py",
"patch": "@@ -15,7 +15,6 @@\n import importlib\n import inspect\n import re\n-import warnings\n from typing import Any, Optional, Union\n \n from packaging import version\n@@ -70,14 +69,9 @@ class PeftAdapterMixin:\n more details about adapters and inje... | 2025-10-02T16:36:57 |
denoland/deno | 6d55117b824ee0a6cfd12ddadf34f570dfc7ebfc | 5f2a56ff5fe46b67a6870b59e407507cde175416 | Revert "fix(ext/node): fix sqlite extension used for testing; ensure … (#31539)
…related tests are actually meaningful (#31455)"
This reverts commit 40cf41275b8e1ec4358d6bca5fdecd104787e5d5.
Started getting random kills on linux arm after this PR landed.
Reverting to see if it will help. | [
{
"path": "Cargo.lock",
"patch": "@@ -8528,13 +8528,6 @@ dependencies = [\n \"winnow 0.6.26\",\n ]\n \n-[[package]]\n-name = \"sqlite_extension_test\"\n-version = \"0.1.0\"\n-dependencies = [\n- \"test_server\",\n-]\n-\n [[package]]\n name = \"stable_deref_trait\"\n version = \"1.2.0\"\n@@ -9354,6 +9347,14... | 2025-12-09T15:35:50 |
ggml-org/llama.cpp | 745cbcf2fe1eb88f8db615ac622f0b944d924ad6 | 1cbd80f8cf80a817715b1ccc5680fe2a3c5172c8 | llama-quant : fix the verification of attention layers for encoder-decoder models (#16023)
Signed-off-by: Jie Fu <jiefu@tencent.com> | [
{
"path": "src/llama-quant.cpp",
"patch": "@@ -725,7 +725,9 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::\n // attention layers have a non-zero number of kv heads\n int32_t n_attn_layer = model.hparams.n_layer - std::count(n_head_kv_iter, n_head_kv_iter ... | 2025-09-17T07:30:55 |
ollama/ollama | 65d93a86b2cf7d94072f34d7f2a2e61c316f4329 | 6d6b0d33219322553c2ac70393e347f1bba61367 | Update modelfile.md (#177)
fix markdown. | [
{
"path": "docs/modelfile.md",
"patch": "@@ -38,7 +38,7 @@ SYSTEM You are Mario from super mario bros, acting as an assistant.\n \n To use this:\n \n-1. Save it as a file (eg. `Modelfile``)\n+1. Save it as a file (eg. `Modelfile`)\n 2. `ollama create NAME -f <location of the file eg. ./Modelfile>'`\n 3. `ol... | 2023-07-22T15:19:30 |
vuejs/vue | 9a742cb423690acb29990bd38f5c17f7b21f12f8 | f4df893828779a74f6379a6068c223a25400828a | only treat binding as domProps on specific elements (fix #4233) | [
{
"path": "flow/compiler.js",
"patch": "@@ -7,7 +7,7 @@ declare type CompilerOptions = {\n directives?: { [key: string]: Function }; // platform specific directives\n isUnaryTag?: (tag: string) => ?boolean; // check if a tag is unary for the platform\n isReservedTag?: (tag: string) => ?boolean; // che... | 2016-11-18T16:36:05 |
huggingface/transformers | e54defcfc265a1cc4f62edbf977747a0a443c812 | b3bd815786c36f4e6c3791fae0a96cac86658b32 | [`Flex Attn`] Fix lse x attention sinks logic (#41249)
fix | [
{
"path": "src/transformers/integrations/flex_attention.py",
"patch": "@@ -282,10 +282,10 @@ def score_mod(score, batch_idx, head_idx, q_idx, kv_idx):\n # On CPU we must skip returning LSE due to a runtime issue; elsewhere, follow PyTorch API and return it\n return_lse = query.device.type != \"cpu\"... | 2025-10-02T15:49:39 |
ggml-org/llama.cpp | 85286f354813056f6c835046c0acfa3bf6ba9432 | d5fabe3682de515fd09d6c981f7a0d1b75614455 | model : add OLMo3 support (#16015)
* Add HF to gguf conversion logic for Olmo3
* Add Olmo3 implementation
* Update rope comment
* Fix indentation
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Apply suggestion from @CISC
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
... | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -6009,9 +6009,34 @@ class SeedOssModel(TextModel):\n \n \n @ModelBase.register(\"Olmo2ForCausalLM\")\n+@ModelBase.register(\"Olmo3ForCausalLM\")\n class Olmo2Model(TextModel):\n model_arch = gguf.MODEL_ARCH.OLMO2\n \n+ def set_gguf_parameters(self):\n+ ... | 2025-09-17T07:01:58 |
denoland/deno | 55a2c7d28b1a4c0714fc5af504a9b7077bfbb3b4 | 05dac153b6eacc7218a4bae1bfdf35509060680b | fix(install): report installed npm packages when not using node_modules, fix double counting (#31536)
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com> | [
{
"path": "cli/tools/installer/mod.rs",
"patch": "@@ -81,7 +81,7 @@ pub struct InstallStats {\n pub downloaded_jsr: DashSet<String>,\n pub reused_jsr: DashSet<String>,\n pub resolved_npm: DashSet<String>,\n- pub downloaded_npm: Count,\n+ pub downloaded_npm: DashSet<String>,\n pub intialized_npm: D... | 2025-12-09T12:20:03 |
vuejs/vue | f4df893828779a74f6379a6068c223a25400828a | 9a659593c48d6e88058515bf407ba61a6d359f79 | fix weex tests for optimizer adjustments | [
{
"path": "test/weex/compiler/class.spec.js",
"patch": "@@ -6,8 +6,8 @@ describe('compile class', () => {\n const { render, staticRenderFns, errors } = compile(`<div class=\"a b c\"></div>`)\n expect(render).not.toBeUndefined()\n expect(staticRenderFns).not.toBeUndefined()\n- expect(staticRen... | 2016-11-17T23:15:10 |
ollama/ollama | 6d6b0d33219322553c2ac70393e347f1bba61367 | 37324a0a007036096755e58e2cd4af030cd0a2c5 | change error handler behavior and fix error when a model isn't found (#173) | [
{
"path": "api/client.go",
"patch": "@@ -210,15 +210,9 @@ func (c *Client) List(ctx context.Context) (*ListResponse, error) {\n \treturn &lr, nil\n }\n \n-type DeleteProgressFunc func(ProgressResponse) error\n-\n-func (c *Client) Delete(ctx context.Context, req *DeleteRequest, fn DeleteProgressFunc) error {... | 2023-07-22T06:02:12 |
huggingface/transformers | b3bd815786c36f4e6c3791fae0a96cac86658b32 | e4930d6bde027ed21fd3dd91dbbfdc95e0221dd4 | Fix mxfp4 dequantization (#41292)
fix | [
{
"path": "src/transformers/integrations/mxfp4.py",
"patch": "@@ -343,7 +343,6 @@ def dequantize(module, param_name, param_value, target_device, dq_param_name, **\n to_contiguous,\n rank,\n device_mesh,\n- set_param=False,\n ... | 2025-10-02T14:47:42 |
ggml-org/llama.cpp | d5fabe3682de515fd09d6c981f7a0d1b75614455 | 8ff206097c2bf3ca1c7aa95f9d6db779fc7bdd68 | CANN: Optimize ggml_cann_set_device (#15935)
* CANN: Fix ggml_cann_set_device to avoid redundant device switches
- Added a check to skip aclrtSetDevice if the current device is already set.
- Prevents unnecessary context switches while keeping thread/device consistency.
* CANN: add device default id | [
{
"path": "ggml/src/ggml-cann/common.h",
"patch": "@@ -526,7 +526,10 @@ struct ggml_backend_cann_context {\n */\n aclrtStream stream(int stream) {\n if (streams[stream] == nullptr) {\n- ggml_cann_set_device(device);\n+ // If the device is not set here, destroying the s... | 2025-09-17T06:33:08 |
denoland/deno | 0bc213b15b819597d935be6cb04cec03659d146b | 63fe08925da3ed245fca9c3935ab40101ec5eb3e | fix(ext/node): prevent cipher operations after finalize (#31533)
Throw a state error for operations if the cipher is already finalized. | [
{
"path": "ext/node/polyfills/internal/crypto/cipher.ts",
"patch": "@@ -184,6 +184,8 @@ export class Cipheriv extends Transform implements Cipher {\n \n #autoPadding = true;\n \n+ #finalized = false;\n+\n constructor(\n cipher: string,\n key: CipherKey,\n@@ -220,6 +222,10 @@ export class Cipher... | 2025-12-09T10:33:37 |
vuejs/vue | 8f7c49c9ba76a288b9eaf8a0bd172fc68218bc6c | f80231773e486f3ef487472f615b785bf7a0b32b | Remove old static style when applying style update (fix #4227) (#4235)
* both static style and stylebinding should be removed
* update test case
* update test case | [
{
"path": "src/platforms/web/runtime/modules/style.js",
"patch": "@@ -42,7 +42,12 @@ function updateStyle (oldVnode: VNodeWithData, vnode: VNodeWithData) {\n \n let cur, name\n const el: any = vnode.elm\n- const oldStyle: any = oldVnode.data.style || {}\n+ const oldStaticStyle: any = oldVnode.data.sta... | 2016-11-17T16:25:54 |
huggingface/transformers | 1d7ebff39807e3ed65674605b8c13954d02bff7f | 9d02602f0f597493257f5a738f3df149c1f93cbf | Fix - remove deprecated args checking in deepspeed intergrations (#41282)
Remove deprecated args checking in deepspeed intergrations
Signed-off-by: nguyen599 <pnvmanh2123@gmail.com>
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com> | [
{
"path": "src/transformers/integrations/deepspeed.py",
"patch": "@@ -356,11 +356,6 @@ def deepspeed_optim_sched(trainer, hf_deepspeed_config, args, num_training_steps\n \n optimizer = None\n if \"optimizer\" in config:\n- if args.optim == \"adafactor\":\n- raise ValueError(\n- ... | 2025-10-02T13:59:50 |
ggml-org/llama.cpp | 3913f8730ec6d6245480affc30ae3049107956f4 | 76888d202ed2b835ae19ea9f9db6baf39e419297 | ggml : fix padding in timestep embedding kernels (#15932)
* ggml : remove adding extra dim timestep embedding
This commit updates the ggml_timestep_embedding function to no longer
add an extra dimension when the specified dimension is odd.
The motivation for this change is that this introduces an unnecessary
dimensi... | [
{
"path": "ggml/src/ggml-cpu/ops.cpp",
"patch": "@@ -8599,7 +8599,6 @@ static void ggml_compute_forward_timestep_embedding_f32(\n }\n if (dim % 2 != 0 && ith == 0) {\n embed_data[2 * half] = 0.f;\n- embed_data[dim] = 0.f;\n }\n }\n }",
"additions": 0,
... | 2025-09-16T13:25:57 |
ollama/ollama | 20a5d99f77ec8fe16aad5f4c9b08cbef7e717ef8 | 3b43cc019a1cf1074eb52eda9c0b9d663ae9f35c | fix vars.First | [
{
"path": "server/images.go",
"patch": "@@ -51,7 +51,7 @@ func (m *Model) Prompt(request api.GenerateRequest) (string, error) {\n \t\tContext []int\n \t}\n \n-\tvars.First = len(vars.Context) == 0\n+\tvars.First = len(request.Context) == 0\n \tvars.System = m.System\n \tvars.Prompt = request.Prompt\n \tvars... | 2023-07-22T03:45:32 |
vuejs/vue | d8309248db4da8e162dc857d92a7ef45bac39e94 | 366202f31668619e6e8d3cb71dc68ec1e9d4f444 | fix eslint | [
{
"path": "src/platforms/weex/runtime/index.js",
"patch": "@@ -1,8 +1,6 @@\n /* @flow */\n \n import Vue from 'core/index'\n-import config from 'core/config'\n-import { noop } from 'shared/util'\n import { patch } from 'weex/runtime/patch'\n import platformDirectives from 'weex/runtime/directives/index'\n i... | 2016-11-16T17:06:27 |
denoland/deno | 80e36b53d386dd9b83e54d2b86e5be6bb10479b6 | 1317a36332cae7fc0712bee17722519919ccae31 | fix(install): rewrite node_modules .bin shim generation for windows to be like npm (#31494)
Previously we were just generating a little cmd file that did `deno run
npm:/..`..., but we were the odd one out in that most other package
managers do .cmd, a sh script, and a powershell script.
This ports the shim logic from... | [
{
"path": "Cargo.lock",
"patch": "@@ -2595,6 +2595,7 @@ dependencies = [\n \"deno_unsync\",\n \"dyn-clone\",\n \"futures\",\n+ \"lazy-regex\",\n \"log\",\n \"once_cell\",\n \"parking_lot\",",
"additions": 1,
"deletions": 0,
"language": "Unknown"
},
{
"path": "libs/npm_installer/Car... | 2025-12-09T08:39:20 |
huggingface/transformers | bc33fd3fc21f9b69e2b48ab8db890435752ac5fd | 639ad8ccd91db497aa2d6f8af7f5218d5adc8b92 | Add processor and intergration test for qwen3vl (#41277)
* support aux loss in qwen3vlmoe
* update qwen3vl processor test!
* add integration tests for qwen3vl-30a3
* remove duplicated decorator
* code clean
* fix consistency
* do not inherit from nn.Linear for better quantization
* pass check | [
{
"path": "src/transformers/models/qwen3_vl/modeling_qwen3_vl.py",
"patch": "@@ -37,7 +37,7 @@\n from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update\n from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel\n from ...processing_utils import Unpack\n-from ...utils impor... | 2025-10-02T12:59:04 |
ggml-org/llama.cpp | f1fbffb5c0b34b2a68febb7da3fd0f8333f1ed4c | 51abc96bdc52ba8cd6ad78dcf12ed9a041d7b442 | fix: apply clang-format to CUDA macros (#16017)
clang-format previously broke long CUDA macros (e.g. __launch_bounds__) into
unreadable line breaks inside template declarations, such as:
template<int D, int ncols, int nwarps, int VKQ_stride,
typename KQ_acc_t, bool use_logit_softcap>
__launch_bound... | [
{
"path": ".clang-format",
"patch": "@@ -22,6 +22,13 @@ AllowShortIfStatementsOnASingleLine: Never\n AllowShortLambdasOnASingleLine: Inline\n AllowShortLoopsOnASingleLine: false\n AlwaysBreakBeforeMultilineStrings: true\n+# Treat CUDA keywords/attributes as \"attribute macros\" and avoid breaking lines insi... | 2025-09-16T06:59:19 |
ollama/ollama | 3b43cc019a1cf1074eb52eda9c0b9d663ae9f35c | b8421dce3d1ed4e3b56aa6818e926ebcfe96c017 | fix extended tag names (#171) | [
{
"path": "server/modelpath.go",
"patch": "@@ -45,7 +45,7 @@ func ParseModelPath(name string) ModelPath {\n \t\treturn ModelPath{}\n \t}\n \n-\tcolonParts := strings.Split(name, \":\")\n+\tcolonParts := strings.Split(slashParts[len(slashParts)-1], \":\")\n \tif len(colonParts) == 2 {\n \t\ttag = colonParts[... | 2023-07-22T03:27:25 |
vuejs/vue | 366202f31668619e6e8d3cb71dc68ec1e9d4f444 | eb92723df54373ebc2aad1790566bdcf058699ae | fix entity-decoder build config | [
{
"path": "build/config.js",
"patch": "@@ -25,6 +25,7 @@ const builds = {\n entry: path.resolve(__dirname, '../src/entries/web-runtime.js'),\n dest: path.resolve(__dirname, '../dist/vue.runtime.js'),\n format: 'umd',\n+ env: 'development',\n banner\n },\n // runtime-only production bu... | 2016-11-16T16:25:40 |
denoland/deno | 1317a36332cae7fc0712bee17722519919ccae31 | b18b8347f7095d85608dda41c47af3c744539787 | fix(bundle): don't use createRequire when targeting browser in Deno.bundle (#31534)
Fixes #31524
Missed a code path in the original fix | [
{
"path": "cli/tools/bundle/provider.rs",
"patch": "@@ -173,7 +173,9 @@ impl BundleProvider for CliBundleProvider {\n super::process_result(\n &result,\n &bundler.cwd,\n- true,\n+ crate::tools::bundle::should_replace_require_shim(\n+ bundl... | 2025-12-09T00:13:35 |
huggingface/transformers | 894a2bdd8c1d5447dcdbf6c5ccaa7cdd7213cae5 | 1cc90695511a4ebb649673081583ac9895b99f23 | Fix pylint generator warnings (#41258)
Fix pylint generator warnings
Signed-off-by: cyy <cyyever@outlook.com> | [
{
"path": "examples/legacy/run_chinese_ref.py",
"patch": "@@ -55,7 +55,7 @@ def get_chinese_word(tokens: list[str]):\n def add_sub_symbol(bert_tokens: list[str], chinese_word_set: set()):\n if not chinese_word_set:\n return bert_tokens\n- max_word_len = max([len(w) for w in chinese_word_set])... | 2025-10-02T12:35:42 |
ggml-org/llama.cpp | 3d4053f77f0f78ee2b791088c02af653ebee42dd | dc381aa9a6dc45f00673471d34b8bddd30e77570 | CUDA: fix im2col_3d to respect non-contiguous inputs (views) (#15956)
* fix im2col_3d to respect non-contiguous inputs (views)
The CUDA 3D im2col kernel computed source addresses assuming compact layout (products of dims), ignoring nb[] strides.
This patch switches im2col_3d source indexing to use true strides de... | [
{
"path": "ggml/src/ggml-cuda/im2col.cu",
"patch": "@@ -122,11 +122,14 @@ static __global__ void im2col_3d_kernel(\n int64_t OH_OW, int64_t KD_KH_KW, int64_t ID_IH_IW, int64_t KH_KW, int64_t IH_IW, int64_t IC_ID_IH_IW,\n int64_t IC_KD_KH_KW, int64_t OW_KD_KH_KW, int64_t OD_OH_OW_IC_KD_KH_KW... | 2025-09-15T22:28:31 |
vuejs/vue | eb92723df54373ebc2aad1790566bdcf058699ae | 02620c928f19716dc16cf2a16c5d2f5ab2378b1d | avoid merging text nodes when the node is a cloned slot node (fix #4209) | [
{
"path": "src/core/vdom/helpers/normalize-children.js",
"patch": "@@ -28,7 +28,9 @@ export function normalizeChildren (\n }\n } else if (c instanceof VNode) {\n if (c.text && last && last.text) {\n- last.text += c.text\n+ if (!last.isCloned) {\n+ last.text... | 2016-11-16T04:07:50 |
denoland/deno | b18b8347f7095d85608dda41c47af3c744539787 | 17cdcc288c741c35465baccae737230eadb574f4 | chore: macos-15-intel runner on ci (#31419)
This includes fixes from
https://github.com/denoland/sui/commit/e229ef18a616e76bee0f225807bc71f8376c3c9b | [
{
"path": ".github/workflows/ci.generate.ts",
"patch": "@@ -12,7 +12,7 @@ const ubuntuX86XlRunner = \"ghcr.io/cirruslabs/ubuntu-runner-amd64:24.04\";\n const ubuntuARMRunner = \"ghcr.io/cirruslabs/ubuntu-runner-arm64:24.04\";\n const windowsX86Runner = \"windows-2022\";\n const windowsX86XlRunner = \"window... | 2025-12-08T23:07:44 |
ggml-org/llama.cpp | 28c39da7c645185ade5436767929d7ec33006033 | 106220562aca42b6738b8f51acfce0db1b8a2fb6 | llama-run: Fix model download on Windows (#15988)
* llama-run: Fix model download on Windows
* fix SSL error (SSL peer certificate or SSH remote key was not OK)
* fix program crash on std::filesystem::rename
* llama-run: create a separate method to utilize RAII
* llama-run: handle rename exception | [
{
"path": "tools/run/run.cpp",
"patch": "@@ -407,6 +407,43 @@ class HttpClient {\n }\n \n std::string output_file_partial;\n+\n+ if (!output_file.empty()) {\n+ output_file_partial = output_file + \".partial\";\n+ }\n+\n+ if (download(url, headers, output_file_... | 2025-09-15T10:08:30 |
vuejs/vue | 02620c928f19716dc16cf2a16c5d2f5ab2378b1d | 83eb63b612432ba0604a9a05f71bafb6ab07c908 | fix weex-compiler externals | [
{
"path": "build/config.js",
"patch": "@@ -76,7 +76,7 @@ const builds = {\n entry: path.resolve(__dirname, '../src/entries/weex-compiler.js'),\n dest: path.resolve(__dirname, '../packages/weex-template-compiler/build.js'),\n format: 'cjs',\n- external: ['entities', 'de-indent']\n+ external... | 2016-11-16T02:43:47 |
huggingface/transformers | 1cc90695511a4ebb649673081583ac9895b99f23 | 4f286fbbf83b7688f4bf299264c1327f5347d3ba | Fix unnecessary single-item container checks (#41279)
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com> | [
{
"path": "examples/legacy/pytorch-lightning/run_ner.py",
"patch": "@@ -72,12 +72,12 @@ def prepare_data(self):\n self.labels,\n args.max_seq_length,\n self.tokenizer,\n- cls_token_at_end=bool(self.config.model_type in [\"xlnet\"... | 2025-10-02T12:35:11 |
ggml-org/llama.cpp | a68f31edd71cc39141113f05f7133a3e9ece8c61 | b8e09f08b9a91c0401bc67d17a17c90756420346 | fix KLD percentile output (#15999)
In `llama-perplexity`, when using `--kl-divergence`, the KL divergence statistics output mistakenly displays the 99th percentile twice. This change fixes that and correctly displays the 90th percentile as originally intended (presumably). | [
{
"path": "tools/perplexity/perplexity.cpp",
"patch": "@@ -1931,7 +1931,7 @@ static void kl_divergence(llama_context * ctx, const common_params & params) {\n LOG(\"Maximum KLD: %10.6f\\n\", kld_values.back());\n LOG(\"99.9%% KLD: %10.6f\\n\", percentile(kld_values, 0.999f));\n LOG(\"99.0%% K... | 2025-09-15T07:54:57 |
denoland/deno | a15cafeb3bf135f3043bc3f78b08b925ffe6a244 | 642f2a46a6d49b2018c042faae5407c328810d9d | feat(ext/web): transferable {Readable,Writable,Transform}Stream (#31126)
https://streams.spec.whatwg.org/#rs-transfer
https://streams.spec.whatwg.org/#ws-transfer
https://streams.spec.whatwg.org/#ts-transfer
Remaining test failures are due to our `DOMException` not correctly
being serializable and can be solved in a ... | [
{
"path": ".gitignore",
"patch": "@@ -45,3 +45,6 @@ Untitled*.ipynb\n /.ms-playwright\n \n **/.claude/settings.local.json\n+\n+# pyenv\n+/.python-version",
"additions": 3,
"deletions": 0,
"language": "Unknown"
},
{
"path": "ext/web/06_streams.js",
"patch": "@@ -60,6 +60,7 @@ const {\... | 2025-12-08T12:31:43 |
huggingface/transformers | f1b64c5b069c24494da845d2995dae0c83fef047 | 2f3e26669202d885fd8bfef5cc27019c355ef916 | Unify is_torchvision_v2_available with is_torchvision_available (#41259)
Fix is_torchvision_v2_available
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com> | [
{
"path": "src/transformers/image_processing_utils_fast.py",
"patch": "@@ -46,7 +46,6 @@\n auto_docstring,\n is_torch_available,\n is_torchvision_available,\n- is_torchvision_v2_available,\n is_vision_available,\n logging,\n )\n@@ -60,14 +59,13 @@\n import torch\n \n if is_torchvi... | 2025-10-02T11:56:37 |
denoland/deno | 642f2a46a6d49b2018c042faae5407c328810d9d | 2353b9ba6af282c63a1e6e88af9a86690f19ec5d | fix(lsp): don't provide organizeImports action when client provides it (#31530) | [
{
"path": "cli/lsp/config.rs",
"patch": "@@ -1208,6 +1208,14 @@ impl Config {\n })()\n .unwrap_or(false)\n }\n+\n+ pub fn client_provided_organize_imports_capable(&self) -> bool {\n+ (|| {\n+ let experimental = self.client_capabilities.experimental.as_ref()?;\n+ experimental.get(\"cl... | 2025-12-08T10:46:22 |
ggml-org/llama.cpp | b8e09f08b9a91c0401bc67d17a17c90756420346 | 6c019cb04e86e2dacfe62ce7666c64e9717dde1f | model : add grok-2 support (#15539)
* add grok-2 support
* type fix
* type fix
* type fix
* "fix" vocab for invalid sequences
* fix expert tensor mapping and spaces in vocab
* add chat template
* fix norm tensor mapping
* rename layer_out_norm to ffn_post_norm
* ensure ffn_post_norm is mapped
* fix experts m... | [
{
"path": "common/common.h",
"patch": "@@ -288,9 +288,9 @@ struct common_params {\n float rope_freq_base = 0.0f; // RoPE base frequency\n float rope_freq_scale = 0.0f; // RoPE frequency scaling factor\n float yarn_ext_factor = -1.0f; // YaRN extrapolation mix factor\n-... | 2025-09-14T21:00:59 |
huggingface/transformers | 2f3e26669202d885fd8bfef5cc27019c355ef916 | 313504bcdde78cde3fd38f19131817dbdfbf7a93 | fix async client for transformers chat (#41255)
* fix-client
* fix | [
{
"path": "src/transformers/commands/chat.py",
"patch": "@@ -687,7 +687,6 @@ async def _inner_run(self):\n \n model = self.args.model_name_or_path + \"@\" + self.args.model_revision\n host = \"http://localhost\" if self.args.host == \"localhost\" else self.args.host\n- client = AsyncI... | 2025-10-02T11:23:37 |
ollama/ollama | 8454f298acc89af4917db0162116fca59a1d9d92 | a3badaf1031a1f74bdcc58a60dfb24bfe3c99451 | fix example `Modelfile`s | [
{
"path": "examples/mario/readme.md",
"patch": "@@ -20,14 +20,8 @@ What the model file looks like:\n ```\n FROM llama2\n PARAMETER temperature 1\n-PROMPT \"\"\"\n-{{- if not .Context }}\n-<<SYS>>\n-You are Mario from super mario bros, acting as an assistant.\n-<</SYS>>\n-\n-{{- end }}\n-[INST] {{ .Prompt }}... | 2023-07-20T22:46:32 |
denoland/deno | 2353b9ba6af282c63a1e6e88af9a86690f19ec5d | 40cf41275b8e1ec4358d6bca5fdecd104787e5d5 | fix(ext/node): implement `DatabaseSync.aggregate()` (#31461)
Implementation is based on
https://github.com/nodejs/node/blob/v24.2.0/src/node_sqlite.cc
It passes all tests from
https://github.com/nodejs/node/blob/v24.2.0/test/parallel/test-sqlite-aggregate-function.mjs
except for two test cases where it uses `mock()` ... | [
{
"path": "ext/node/ops/sqlite/database.rs",
"patch": "@@ -25,6 +25,7 @@ use deno_permissions::PermissionsContainer;\n use rusqlite::ffi as libsqlite3_sys;\n use rusqlite::ffi::SQLITE_DBCONFIG_DQS_DDL;\n use rusqlite::ffi::SQLITE_DBCONFIG_DQS_DML;\n+use rusqlite::ffi::sqlite3_create_window_function;\n use r... | 2025-12-08T10:34:31 |
ggml-org/llama.cpp | 0fa154e3502e940df914f03b41475a2b80b985b0 | 261e6a20ffdb79c4875e674b4f6b514bc73cff8f | rocm.Dockerfile: added gfx1200,gfx1201 architectures to support AMD Radeon RX 9000 series (#15994)
* rocm.Dockerfile: added gfx1200,gfx1201 architectures to support AMD Radeon RX 9000 series
https://rocm.docs.amd.com/projects/install-on-linux/en/docs-6.4.1/reference/system-requirements.html#rdna-os
states the Radeo... | [
{
"path": ".devops/rocm.Dockerfile",
"patch": "@@ -4,7 +4,7 @@ ARG UBUNTU_VERSION=24.04\n ARG ROCM_VERSION=6.4\n ARG AMDGPU_VERSION=6.4\n \n-# Target the CUDA build image\n+# Target the ROCm build image\n ARG BASE_ROCM_DEV_CONTAINER=rocm/dev-ubuntu-${UBUNTU_VERSION}:${ROCM_VERSION}-complete\n \n ### Build i... | 2025-09-14T18:43:54 |
vuejs/vue | 438b51793f9347407ffc71ea2b432ed7af59d258 | 8c0fac699a48ec62b25a94b4083ea1d75c52c236 | fix style merging assertion in IE | [
{
"path": "test/unit/features/directives/style.spec.js",
"patch": "@@ -195,7 +195,12 @@ describe('Directive v-bind:style', () => {\n }).$mount()\n const style = vm.$el.style\n const child = vm.$children[0]\n- expect(style.cssText.replace(/\\s/g, '')).toBe('margin-right:20px;margin-left:16px;t... | 2016-11-15T20:53:26 |
denoland/deno | 40cf41275b8e1ec4358d6bca5fdecd104787e5d5 | b8b549d574eceed3275923339f402a4b3df5385e | fix(ext/node): fix sqlite extension used for testing; ensure related tests are actually meaningful (#31455)
The SQLite extension-loading tests, prior to this PR, were
short-circuiting due to an incorrect search path, and not actually
verifying that extensions loaded. When this invalid path was fixed, the
result was a ... | [
{
"path": "Cargo.lock",
"patch": "@@ -8527,6 +8527,13 @@ dependencies = [\n \"winnow 0.6.26\",\n ]\n \n+[[package]]\n+name = \"sqlite_extension_test\"\n+version = \"0.1.0\"\n+dependencies = [\n+ \"test_server\",\n+]\n+\n [[package]]\n name = \"stable_deref_trait\"\n version = \"1.2.0\"\n@@ -9346,14 +9353,6... | 2025-12-08T10:30:54 |
ggml-org/llama.cpp | 261e6a20ffdb79c4875e674b4f6b514bc73cff8f | a0e13dcbe5bae7025660349ef3e4ead060e507f2 | Vulkan: Clean up mul_mm shader (#15987)
* vulkan: move mul_mm dequantization steps into a separate file and functions
* improve mul_mm vector load code
* fix debug mode issues and warnings | [
{
"path": "ggml/src/ggml-vulkan/ggml-vulkan.cpp",
"patch": "@@ -1231,8 +1231,6 @@ static std::string format_size(size_t size) {\n return oss.str();\n }\n \n-static std::mutex log_mutex;\n-\n class vk_memory_logger {\n public:\n vk_memory_logger(): total_device(0), total_host(0) {}\n@@ -1422,6 +1420,... | 2025-09-14T14:56:28 |
huggingface/transformers | 7938e91faabb051f3a001cd39c173d4697c2d81c | e6a8e7debed69d4cf5d493158f3f347931c24c9a | MoE + vllm = 😻 (#40132)
* update modeling mixtral
* oups[13;2u
* fix
* better naming?
* compute softmax and top_k inside the experts
* update minamax as well
* models that will need an update
* more models that need a fix
* stash
* fix mixtral
* update olmoe
* update
* update
* current changes
* nits
... | [
{
"path": "docs/source/en/model_doc/switch_transformers.md",
"patch": "@@ -105,7 +105,6 @@ print(tokenizer.decode(outputs[0]))\n ## SwitchTransformersTop1Router\n \n [[autodoc]] SwitchTransformersTop1Router\n- - _compute_router_probabilities\n - forward\n \n ## SwitchTransformersSparseMLP",
"addi... | 2025-10-02T10:12:44 |
ollama/ollama | 6cea2061ec97130140c06f8bacc769a2c984e48e | 2832801c2a76a3c091c06f34fc3379402ca24b55 | windows: fix model pulling | [
{
"path": "server/images.go",
"patch": "@@ -854,6 +854,10 @@ func downloadBlob(mp ModelPath, digest string, username, password string, fn fun\n \t\t})\n \n \t\tif completed >= total {\n+\t\t\tif err := out.Close(); err != nil {\n+\t\t\t\treturn err\n+\t\t\t}\n+\n \t\t\tif err := os.Rename(fp+\"-partial\", f... | 2023-07-20T18:23:43 |
vuejs/vue | 8c0fac699a48ec62b25a94b4083ea1d75c52c236 | ef4fc1c75221bde2f4df915bba7d3585cad37c5c | fix hydration node match assertion for elements with camelCase names (fix #4195) | [
{
"path": "src/core/vdom/patch.js",
"patch": "@@ -446,7 +446,7 @@ export function createPatchFunction (backend) {\n if (vnode.tag) {\n return (\n vnode.tag.indexOf('vue-component') === 0 ||\n- vnode.tag === nodeOps.tagName(node).toLowerCase()\n+ vnode.tag.toLowerCase() === no... | 2016-11-15T20:21:59 |
denoland/deno | b8b549d574eceed3275923339f402a4b3df5385e | 61cf9333a914fa3eb1d5295f01e2ccb12f08c3cb | fix(node): missing `statfs` export from `node:fs/promises` (#31528)
We already had the functionality, we just didn't export it.
Fixes https://github.com/denoland/deno/issues/31526 | [
{
"path": "ext/node/polyfills/fs/promises.ts",
"patch": "@@ -16,6 +16,7 @@ export const readlink = fsPromises.readlink;\n export const symlink = fsPromises.symlink;\n export const lstat = fsPromises.lstat;\n export const stat = fsPromises.stat;\n+export const statfs = fsPromises.statfs;\n export const fstat... | 2025-12-08T09:45:01 |
ggml-org/llama.cpp | a0e13dcbe5bae7025660349ef3e4ead060e507f2 | a14bd350141fb42b8bf2dd2342cebc27bfdce399 | build: fix the build failures of Windows HIP release job (#15984)
* build: fix the cache keys for Windows HIP release job
Update the cache keys to include the HIP SDK version, preventing the
use of outdated ROCm installation caches.
* build: sync changes from release.yml to build.yml
- Update HIP SDK version to 25.... | [
{
"path": ".github/workflows/build.yml",
"patch": "@@ -127,7 +127,8 @@ jobs:\n -DCMAKE_BUILD_RPATH=\"@loader_path\" \\\n -DLLAMA_FATAL_WARNINGS=ON \\\n -DGGML_METAL=OFF \\\n- -DGGML_RPC=ON\n+ -DGGML_RPC=ON \\\n+ -DCMAKE_OSX_DEPLOYMENT_TARG... | 2025-09-14T14:20:35 |
huggingface/transformers | e6a8e7debed69d4cf5d493158f3f347931c24c9a | 30b79effb5d13c9319c1a32aa239c1cc238f67cb | Fix binding of video frames to video placeholder in `InternVL` model (#41237)
* Fix binding video frames to video placeholder in prompt
Signed-off-by: Daniel Bershatsky <daniel.bershatsky@gmail.com>
* Add test on binding video frames to prompt
Signed-off-by: Daniel Bershatsky <daniel.bershatsky@gmail.com>
* Fix co... | [
{
"path": "src/transformers/models/internvl/processing_internvl.py",
"patch": "@@ -40,7 +40,9 @@ class InternVLProcessorKwargs(ProcessingKwargs, total=False):\n \"images_kwargs\": {\n \"crop_to_patches\": True,\n },\n- \"videos_kwargs\": {},\n+ \"videos_kwargs\": {\... | 2025-10-02T09:43:35 |
vuejs/vue | ef4fc1c75221bde2f4df915bba7d3585cad37c5c | 79cc7bc97a6aa527057756e2e441cad399b299b1 | fix eslint | [
{
"path": "src/core/util/options.js",
"patch": "@@ -6,7 +6,6 @@ import { warn } from './debug'\n import { set } from '../observer/index'\n import {\n extend,\n- isObject,\n isPlainObject,\n hasOwn,\n camelize,",
"additions": 0,
"deletions": 1,
"language": "JavaScript"
}
] | 2016-11-15T20:21:34 |
denoland/deno | 99c4d22b8574557921e45466df1f6d4f569bee8b | c27d8f047bec19a175e8c5708e865fba85189023 | Revert "fix(resolver): fall back to execution if can't resolve types" (#31513)
Reverts denoland/deno#31507 | [
{
"path": "libs/resolver/npm/mod.rs",
"patch": "@@ -513,21 +513,6 @@ impl<\n match resolution_result {\n Ok(url) => Ok(url),\n Err(err) => {\n- if err.as_types_not_found().is_some() {\n- // failed to find types, just fall back to normal resolution\n- debug_assert_eq!... | 2025-12-07T15:46:49 |
ggml-org/llama.cpp | a14bd350141fb42b8bf2dd2342cebc27bfdce399 | 918b26f197f55d5d562446dfc876d0e637929d07 | metal : fix kernel requirements (#15983)
* metal : fix kernel requirements
ggml-ci
* cont : fix supports_op
* cont : fix supports_op for ARGMAX | [
{
"path": "ggml/src/ggml-metal/ggml-metal.m",
"patch": "@@ -1219,10 +1219,10 @@ @implementation GGMLMetalClass\n GGML_METAL_ADD_KERNEL(GGML_METAL_KERNEL_TYPE_SET_ROWS_IQ4_NL, set_rows_iq4_nl, true);\n GGML_METAL_ADD_KERNEL(GGML_METAL_KERNEL_TYPE_L2_NORM, ... | 2025-09-14T12:33:22 |
huggingface/transformers | 30b79effb5d13c9319c1a32aa239c1cc238f67cb | aabf0a03cb5ebf34bf17cf157c88a34a76cb8a86 | Remove SageMakerTrainer (#41267)
* Remove SageMakerTrainer
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com>
* More removal
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com>
* More fixes
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com>
---------
Signed-off-by: Yuanyuan Chen <cyyever@outlook.com> | [
{
"path": "src/transformers/__init__.py",
"patch": "@@ -471,7 +471,6 @@\n \"prune_layer\",\n \"infer_device\",\n ]\n- _import_structure[\"sagemaker\"] = []\n _import_structure[\"time_series_utils\"] = []\n _import_structure[\"trainer\"] = [\"Trainer\"]\n _import_structure[... | 2025-10-02T09:16:32 |
vuejs/vue | 79cc7bc97a6aa527057756e2e441cad399b299b1 | ab277adfbe2ac7776d5d29d82a267a6a06bc8361 | improve mergeData for edge cases (fix #4191) | [
{
"path": "src/core/util/options.js",
"patch": "@@ -40,13 +40,16 @@ if (process.env.NODE_ENV !== 'production') {\n * Helper that recursively merges two data objects together.\n */\n function mergeData (to: Object, from: ?Object): Object {\n+ if (!from) return to\n let key, toVal, fromVal\n- for (key i... | 2016-11-15T20:15:04 |
ggml-org/llama.cpp | 918b26f197f55d5d562446dfc876d0e637929d07 | 9ecb88434644c865232bb665d2f6f05049fc6456 | rpc : fix regression when --device is used (#15981)
Fix regression introduced with commit 50f4281a6 | [
{
"path": "tools/rpc/rpc-server.cpp",
"patch": "@@ -227,7 +227,9 @@ static ggml_backend_t create_backend(const rpc_server_params & params) {\n }\n }\n \n- backend = ggml_backend_init_best();\n+ if (!backend) {\n+ backend = ggml_backend_init_best();\n+ }\n \n if (backend) {\n ... | 2025-09-14T09:28:18 |
denoland/deno | c27d8f047bec19a175e8c5708e865fba85189023 | bc9d356059abf4bb04c9849edbfcc53ebcd363ed | fix: make '@types' package resolution more consistent (#31512)
Shifts `@types` package resolution down into node_resolver so it's
always done. We were being inconsistent. | [
{
"path": "Cargo.lock",
"patch": "@@ -2520,9 +2520,9 @@ dependencies = [\n \n [[package]]\n name = \"deno_npm\"\n-version = \"0.42.1\"\n+version = \"0.42.2\"\n source = \"registry+https://github.com/rust-lang/crates.io-index\"\n-checksum = \"7676aeffeeefb7ef4343a5f5c04e6bf97d5b24038d04c119fb3891fa56a1d146\"... | 2025-12-06T16:59:06 |
huggingface/transformers | aabf0a03cb5ebf34bf17cf157c88a34a76cb8a86 | bcdd5532bfd48660c029a0ddb8f9c76cb02039b2 | Fix multi-video timestamp bug in Qwen-3-VL and GLM4V (#41229)
* fix multi-video timestamp bug in qwen3vl,glm4v
* run make fix-copies to sync modular files
* run make fix-copies to sync modular files
---------
Co-authored-by: UBT <daqin.luo@ubtrobot.com> | [
{
"path": "src/transformers/models/glm4v/modular_glm4v.py",
"patch": "@@ -1639,7 +1639,7 @@ def __call__(\n num_frames = video_grid_thw[video_index][0]\n video_structure = \"\"\n \n- metadata = video_metadata[i]\n+ metadata = vide... | 2025-10-02T09:15:57 |
ollama/ollama | 1f27d7f1b8bff58c63753ae2bb29de38454928d6 | 00aaa0590125a07dfd27b2b68fc44e0a99842d2b | fix stream errors | [
{
"path": "api/client.go",
"patch": "@@ -131,6 +131,10 @@ func (c *Client) stream(ctx context.Context, method, path string, data any, fn f\n \t\t\treturn fmt.Errorf(\"unmarshal: %w\", err)\n \t\t}\n \n+\t\tif errorResponse.Error != \"\" {\n+\t\t\treturn fmt.Errorf(\"stream: %s\", errorResponse.Error)\n+\t\t... | 2023-07-20T19:12:08 |
vuejs/vue | ab277adfbe2ac7776d5d29d82a267a6a06bc8361 | 3c78553330ffea8543479e9aac99ac8d1280c431 | work around IE textarea placeholder innerHTML/outerHTML bug (fix #4098) | [
{
"path": "src/compiler/parser/index.js",
"patch": "@@ -211,6 +211,13 @@ export function parse (\n }\n return\n }\n+ // IE textarea placeholder bug\n+ /* istanbul ignore if */\n+ if (options.isIE &&\n+ currentParent.tag === 'textarea' &&\n+ currentParen... | 2016-11-15T18:41:57 |
ggml-org/llama.cpp | 6380d6a3e709cb02a8695afdd96b40e674477332 | aa0c461efe3603639af1a1defed2438d9c16ca0f | ggml-zdnn: rm user mapped buffers (#15965)
* ggml-zdnn: rm user mapped buffers
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
* ggml-zdnn: rm dead code
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
* ggml-zdnn: attempt to fix missing extra data buffer free
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
---------
... | [
{
"path": "ggml/src/ggml-zdnn/ggml-zdnn.cpp",
"patch": "@@ -127,11 +127,6 @@ static void ggml_zdnn_mul_mat_op(ggml_backend_zdnn_context * ctx, const ggml_ten\n const int64_t output_rows = ne1;\n const int64_t output_cols = ne0;\n \n- // TODO: Weights are somehow not going through `ggml_backend_zd... | 2025-09-14T05:37:03 |
denoland/deno | bc9d356059abf4bb04c9849edbfcc53ebcd363ed | 8c46ac3228762fb442b0080c7eb81e5e918543bb | fix(node/net): return string `family` in `server.address()` (#31465)
## Summary
Aligns Deno's Node.js compatibility layer (`node:net`, `node:http`,
`node:https`,
`node:http2`, `node:dns`) with Node.js v18.4.0+ behavior by returning
the `family`
property as a string (`"IPv4"` or `"IPv6"`) rather than a number in
`serv... | [
{
"path": "ext/node/polyfills/dns.ts",
"patch": "@@ -234,8 +234,19 @@ export function lookup(\n }\n \n if (options?.family != null) {\n- validateOneOf(options.family, \"options.family\", validFamilies);\n- family = options.family;\n+ // Accept both numeric (0, 4, 6) and string ('IPv4'... | 2025-12-06T03:51:55 |
huggingface/transformers | 55d63e86eac12c127ad420ec5b18c99eaf75ae62 | 522b79a346c4a188fd16548a7e0bc0ad15b86e5a | fix asr pipeline ut failures (#41275)
* fix asr pipeline ut failures
Signed-off-by: Yao, Matrix <matrix.yao@intel.com>
* make style
Signed-off-by: Yao, Matrix <matrix.yao@intel.com>
---------
Signed-off-by: Yao, Matrix <matrix.yao@intel.com> | [
{
"path": "tests/pipelines/test_pipelines_automatic_speech_recognition.py",
"patch": "@@ -615,7 +615,8 @@ def test_torch_whisper_batched(self):\n {\"text\": \" Nor is Mr. Quilters' manner less interesting than his matter.\"},\n ]\n \n- output = speech_recognizer(ds[\"audio\"], bat... | 2025-10-02T08:32:03 |
vuejs/vue | 3c78553330ffea8543479e9aac99ac8d1280c431 | 3e8ac270a8d62630cae4191e016602c36b2b6beb | warn empty or missing template element (fix #4171) | [
{
"path": "src/core/instance/lifecycle.js",
"patch": "@@ -44,7 +44,7 @@ export function lifecycleMixin (Vue: Class<Component>) {\n vm.$options.render = emptyVNode\n if (process.env.NODE_ENV !== 'production') {\n /* istanbul ignore if */\n- if (vm.$options.template) {\n+ if ... | 2016-11-15T17:29:39 |
ollama/ollama | 09dc6273e3404cf6a3dfad3810a081b4226667bb | ebaa33ac2817e36b37f3c789ee30f4cf9cec957e | suppress error when running list before pulling image | [
{
"path": "server/routes.go",
"patch": "@@ -2,6 +2,7 @@ package server\n \n import (\n \t\"encoding/json\"\n+\t\"errors\"\n \t\"io\"\n \t\"log\"\n \t\"net\"\n@@ -163,6 +164,10 @@ func list(c *gin.Context) {\n \t}\n \terr = filepath.Walk(fp, func(path string, info os.FileInfo, err error) error {\n \t\tif err... | 2023-07-20T18:53:09 |
ggml-org/llama.cpp | aa0c461efe3603639af1a1defed2438d9c16ca0f | b9c9c9f789cd57fb0b28b25223305613cd90fa10 | vulkan: fix failing dequant shaders (#15862)
* vulkan: fix failing dequant shaders
* add missing const | [
{
"path": "ggml/src/ggml-vulkan/vulkan-shaders/dequant_iq2_s.comp",
"patch": "@@ -29,7 +29,7 @@ void main() {\n uint qs = data_a[ib].qs[4 * ib32 + l];\n const uint8_t sign = data_a[ib].qs[QUANT_K / 8 + 4 * ib32 + l];\n qs |= (qh << (8 - 2 * l)) & 0x300;\n- const uvec2 grid = i... | 2025-09-13T15:29:43 |
denoland/deno | 8c46ac3228762fb442b0080c7eb81e5e918543bb | 3a4b9de81ef0a83652729320cfc750f94ccab924 | fix(check): move types not found error to tsc diagnostic (#31509)
Closes https://github.com/denoland/deno/issues/31352 | [
{
"path": "cli/graph_util.rs",
"patch": "@@ -297,9 +297,9 @@ pub fn resolution_error_for_tsc_diagnostic(\n | NodeJsErrorCode::ERR_UNSUPPORTED_ESM_URL_SCHEME\n | NodeJsErrorCode::ERR_INVALID_FILE_URL_PATH\n | NodeJsErrorCode::ERR_PACKAGE_IMPORT_NOT_DEFINED\n- | NodeJsErrorCode::ERR_PAC... | 2025-12-06T02:57:45 |
huggingface/transformers | 522b79a346c4a188fd16548a7e0bc0ad15b86e5a | 9f2d5666f8fda5b647e5f64dfc8ba1edd7a87a1e | add more activation kernels, follow up (#40944)
* add more activation kernels
* fixing style
* fix version | [
{
"path": "src/transformers/activations.py",
"patch": "@@ -12,6 +12,7 @@\n # See the License for the specific language governing permissions and\n # limitations under the License.\n \n+import functools\n import math\n from collections import OrderedDict\n \n@@ -26,7 +27,8 @@\n logger = logging.get_logger(__... | 2025-10-02T06:45:05 |
ollama/ollama | ebaa33ac2817e36b37f3c789ee30f4cf9cec957e | 6a19724d5f48b593a2519293b8485d528796ef7c | display gin api errors in cli | [
{
"path": "api/client.go",
"patch": "@@ -27,7 +27,7 @@ func checkError(resp *http.Response, body []byte) error {\n \terr := json.Unmarshal(body, &apiError)\n \tif err != nil {\n \t\t// Use the full body as the message if we fail to decode a response.\n-\t\tapiError.Message = string(body)\n+\t\tapiError.Erro... | 2023-07-20T18:45:12 |
vuejs/vue | 3e8ac270a8d62630cae4191e016602c36b2b6beb | 77497931e00753e109d8a72549c00b0bccf45461 | fix .trim modifier when v-model is used on custom component (fix #4204) | [
{
"path": "src/platforms/web/compiler/directives/model.js",
"patch": "@@ -123,7 +123,7 @@ function genDefaultModel (\n \n let valueExpression = isNative\n ? `$event.target.value${trim ? '.trim()' : ''}`\n- : `$event`\n+ : trim ? `(typeof $event === 'string' ? $event.trim() : $event)` : `$event`\... | 2016-11-15T16:05:08 |
ggml-org/llama.cpp | b9c9c9f789cd57fb0b28b25223305613cd90fa10 | 50f4281a6f5c3a5d68bdeb12f904fa01e0e2ba91 | vulkan: initialize vulkan-hpp to allow using extension function pointers (#15705)
Use this to query register count for shader compiles on NVIDIA. Currently
this is only for performance debug, but it could eventually be used in some
heuristics like split_k. | [
{
"path": "ggml/src/ggml-vulkan/ggml-vulkan.cpp",
"patch": "@@ -5,8 +5,14 @@\n #include \"ggml-cpu.h\"\n #endif\n \n+// See https://github.com/KhronosGroup/Vulkan-Hpp?tab=readme-ov-file#extensions--per-device-function-pointers-\n+#define VULKAN_HPP_DISPATCH_LOADER_DYNAMIC 1\n+\n #include <vulkan/vulkan.hpp>... | 2025-09-13T15:23:30 |
huggingface/transformers | c4b505d0f7e696bf67f999d6ba958931322f2def | 01c9e1ba683b3e50d7c76bf92f2d470759fd5e81 | Don't convert to `safetensors` on the fly if the call is from testing (#41194)
* don't convert
* disable
* Update src/transformers/modeling_utils.py
Co-authored-by: Cyril Vallez <cyril.vallez@huggingface.co>
* fix
* disable
* disable
* disable
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.co... | [
{
"path": ".circleci/create_circleci_config.py",
"patch": "@@ -29,6 +29,7 @@\n \"RUN_PIPELINE_TESTS\": False,\n # will be adjust in `CircleCIJob.to_dict`.\n \"RUN_FLAKY\": True,\n+ \"DISABLE_SAFETENSORS_CONVERSION\": True,\n }\n # Disable the use of {\"s\": None} as the output is way too long... | 2025-10-01T15:46:21 |
denoland/deno | 3a4b9de81ef0a83652729320cfc750f94ccab924 | 3089aa8690af8dfaa6f24051800fa116a20b3ee3 | fix: change errors message suggesting to run approve-scripts (#31493)
Follow up to https://github.com/denoland/deno/pull/31472
Before:
```
$ deno install
...
╭ Warning
│
│ Ignored build scripts for packages:
│ npm:msgpackr-extract@3.0.3
│ npm:@parcel/watcher@2.5.1
│ npm:lmdb@2.8.5
│ npm:@swc/core@1.12.0
│
│ Run... | [
{
"path": "cli/tools/pm/approve_scripts.rs",
"patch": "@@ -126,11 +126,21 @@ pub async fn approve_scripts(\n config_updater.set_allow_scripts_value(allow_scripts_value);\n config_updater.commit()?;\n \n- for req in approvals {\n- log::info!(\"Approved {}\", colors::green(format!(\"npm:{req}\")));\n-... | 2025-12-05T21:19:29 |
vuejs/vue | 77497931e00753e109d8a72549c00b0bccf45461 | cccc277dd933bf06f2186e894130f5b78c4d448e | fix v-once inside v-for and v-once with v-if (#4200) | [
{
"path": "src/compiler/codegen/index.js",
"patch": "@@ -85,7 +85,9 @@ function genStatic (el: ASTElement): string {\n // v-once\n function genOnce (el: ASTElement): string {\n el.onceProcessed = true\n- if (el.staticInFor) {\n+ if (el.if && !el.ifProcessed) {\n+ return genIf(el)\n+ } else if (el.st... | 2016-11-14T21:20:21 |
ollama/ollama | 99ccf0c5d39420f3f1447536fa62ba3e624257db | d59b164fa2a336e16bd89024959f0a79350c13bd | fix broken link in `README.md` | [
{
"path": "README.md",
"patch": "@@ -82,7 +82,7 @@ ollama run mario\n Hello! It's your friend Mario.\n ```\n \n-For more info on `Modelfile` syntax see [this doc](./docs/modelfile).\n+For more examples, see the [examples](./examples) directory.\n \n ### Pull a model from the registry\n ",
"additions": 1... | 2023-07-20T09:15:11 |
ggml-org/llama.cpp | f161463a54d9f93d41246286aa4a9569a91d804d | 84d7b2fca11d1be118ce776f6d72a486c4883b74 | metal : allow ops to run concurrently (#15929)
* metal : run graphs ops concurrently
ggml-ci
* cont : add flags for debugging and disabling concurrency
ggml-ci
* cont : refactor and handle fusing
ggml-ci
* cont : simplify - no need to use GPU address
ggml-ci
* cont : prepare mem ranges for reuse + add ggml-met... | [
{
"path": "ggml/src/ggml-metal/CMakeLists.txt",
"patch": "@@ -6,6 +6,7 @@ message(STATUS \"Metal framework found\")\n \n ggml_add_backend_library(ggml-metal\n ggml-metal.m\n+ ggml-metal-common.cpp\n )\n \n target_link_libraries(ggml-me... | 2025-09-13T10:54:28 |
huggingface/transformers | 01c9e1ba683b3e50d7c76bf92f2d470759fd5e81 | 025531981cd9cfa538655987d5237decacd91c48 | [t5gemma] fix `get_text_config` and related fixes (#40939)
* tmp commit
* t5gemma fixes | [
{
"path": "src/transformers/cache_utils.py",
"patch": "@@ -395,7 +395,12 @@ def update(\n if not self.is_initialized:\n self.lazy_initialization(key_states)\n \n- cache_position = cache_kwargs.get(\"cache_position\")\n+ # Some old models give None for `cache_position` or ev... | 2025-10-01T14:55:26 |
denoland/deno | a39796500df53799c72e83d7af6a3603625e263a | dfe26dadb0018f07f5a6086e10e4f6562b3613d0 | fix(resolver): fall back to execution if can't resolve types (#31507)
Fixes https://github.com/denoland/deno/issues/31352
This matches the behavior of the other node_modules_dir modes more
closely as well as tsc's behavior. Failing to find types should not be
an error | [
{
"path": "libs/resolver/npm/mod.rs",
"patch": "@@ -526,6 +526,19 @@ impl<\n {\n return Ok(resolved);\n }\n+ // failed to find types, just fall back to normal resolution\n+ debug_assert_eq!(resolution_kind, NodeResolutionKind::Types);\n+ if let Ok(r... | 2025-12-05T20:34:02 |
vuejs/vue | cccc277dd933bf06f2186e894130f5b78c4d448e | 4e38b54274a664db1e101a27a6d3db9860fa91c1 | fix eslint | [
{
"path": "test/weex/compiler/compile.spec.js",
"patch": "@@ -25,6 +25,7 @@ describe('compile class', () => {\n it('should compile data bindings with children', () => {\n const { render, staticRenderFns, errors } = compile(`<foo :a=\"b\"><text>Hello</text></foo>`)\n expect(render).toEqual(`with(th... | 2016-11-13T19:07:34 |
ollama/ollama | 3b135ac96326a215c2406156d3280424c381af69 | e6bae8d916d976d227489fe0daa57586472c6c4e | parser: fix case where multi line string termination error wouldnt show | [
{
"path": "parser/parser.go",
"patch": "@@ -62,10 +62,6 @@ func Parse(reader io.Reader) ([]Command, error) {\n }\n \n func scanModelfile(data []byte, atEOF bool) (advance int, token []byte, err error) {\n-\tif atEOF || len(data) == 0 {\n-\t\treturn 0, nil, nil\n-\t}\n-\n \tnewline := bytes.IndexByte(data, '... | 2023-07-20T07:43:22 |
ggml-org/llama.cpp | 84d7b2fca11d1be118ce776f6d72a486c4883b74 | 40be51152d4dc2d47444a4ed378285139859895b | metal : fix memory leaks (#15962)
ggml-ci | [
{
"path": "ggml/src/ggml-metal/ggml-metal.m",
"patch": "@@ -1361,7 +1361,6 @@ @implementation GGMLMetalClass\n GGML_METAL_ADD_KERNEL(GGML_METAL_KERNEL_TYPE_MUL_MM_Q5_1_F32, mul_mm_q5_1_f32, has_simdgroup_mm);\n GGML_METAL_ADD_KERNEL(GGML_METAL_KERNEL_TYPE_MUL_... | 2025-09-13T09:45:04 |
huggingface/transformers | 025531981cd9cfa538655987d5237decacd91c48 | 32567739740da86ddf96c60a23cf2d0494ce4145 | [`FA3`] Fix masking and loading logic in same process (#41217)
fix loading and fa3 masking | [
{
"path": "src/transformers/integrations/hub_kernels.py",
"patch": "@@ -191,7 +191,7 @@ def load_and_register_kernel(attn_implementation: str) -> None:\n if attention_wrapper is None:\n attention_wrapper = flash_attention_forward\n kernel_function = partial(attention_wrapper, imp... | 2025-10-01T14:36:12 |
denoland/deno | dd0aa99f18822728c80011bb51ae04a9d5ec1b3c | e5f9b441ef9e2ba75306679068ad292a29f690ef | fix: fix npx running in test, make approve-scripts test less flaky (#31501)
It was doing `npx cowsay` but cowsay wasn't listed as a dependency, so
it was falling back to actual npx.
I also found that the cowsay part sometimes overlaps with the progress
bar, so checking for 'install' is more reliable | [
{
"path": "tests/integration/pm_tests.rs",
"patch": "@@ -178,9 +178,7 @@ fn approve_scripts_basic() {\n pty.write_line(\" \");\n pty.write_line(\"\\r\\n\");\n pty.expect(\"Approved npm:@denotest/node-lifecycle-scripts@1.0.0\");\n- pty.expect(\n- \"@denotest/node-lifecycle-scrip... | 2025-12-05T08:20:01 |
vuejs/vue | 4e38b54274a664db1e101a27a6d3db9860fa91c1 | 46b3bcd707addd0e503cc44bf7914b0281780d5c | fix weex tests for component slot static check change | [
{
"path": "test/weex/compiler/compile.spec.js",
"patch": "@@ -22,10 +22,9 @@ describe('compile class', () => {\n expect(errors).toEqual([])\n })\n \n- it('should compile data bindings with static children', () => {\n+ it('should compile data bindings with children', () => {\n const { render, sta... | 2016-11-13T19:04:02 |
ollama/ollama | ada0add89be0da299be3a59e52071be4ea6669e8 | 75e508e1d655cd4889c5cd3c77bbb933025c2f73 | fix `llama` library templates | [
{
"path": "library/modelfiles/llama2",
"patch": "@@ -1,15 +1,19 @@\n FROM ../models/llama-2-7b-chat.ggmlv3.q4_0.bin\n \n-PROMPT \"\"\"\n-{{- if not .Context }}\n+TEMPLATE \"\"\"\n+{{- if .First }}\n <<SYS>>\n-You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while b... | 2023-07-20T06:53:40 |
ggml-org/llama.cpp | 40be51152d4dc2d47444a4ed378285139859895b | 4bf5549269d99bac936a65892da2312cc71b2421 | ggml-zdnn: fix #15414, activate FP16 and BF16 acceleration and incorrect zTensor free (#15839) | [
{
"path": "docs/build-s390x.md",
"patch": "@@ -241,8 +241,8 @@ IBM VXE/VXE2 SIMD acceleration depends on the BLAS implementation. It is strongl\n | | VX/VXE/VXE2 | zDNN | Spyre |\n |------------|-------------|------|-------|\n | FP32 | ✅ | ✅ | ❓ |\n-| FP16 | ✅ ... | 2025-09-12T18:39:52 |
huggingface/transformers | 32567739740da86ddf96c60a23cf2d0494ce4145 | d848a3953aeb7debdbfbc6bfbf788c528cd40e46 | FP-Quant NVFP4 and Python 3.9 support (#39876)
* quartet
* quartet qat -> quartet
* format
* bf16 backward
* interfaces
* forward_method
* quartet -> fp_quant
* style
* List -> list
* list typing
* fixed format and annotations
* test_fp_quant
* docstrings and default dtypes
* better docstring and removed ... | [
{
"path": "docker/transformers-quantization-latest-gpu/Dockerfile",
"patch": "@@ -82,6 +82,9 @@ RUN python3 -m pip uninstall -y flash-attn\n # this line must be added in order for python to be aware of transformers.\n RUN cd transformers && python3 setup.py develop\n \n+# Add fp-quant for quantization testi... | 2025-10-01T13:58:22 |
denoland/deno | 49fa02d98859f467e0216bfdb79417aace90f433 | 54412e4e8fd1fe3dadb9dfb9276a9dfccca90fd1 | fix(ext/node): `url.domainToASCII` returns empty string for invalid domains (#31219)
## Summary
This PR fixes the behavior of `url.domainToASCII` to match Node.js by
returning an empty string when an invalid domain is passed, instead of
throwing an error.
## Changes
- Modified `op_node_idna_domain_to_ascii` in `ext... | [
{
"path": "ext/node/ops/idna.rs",
"patch": "@@ -114,12 +114,12 @@ pub fn op_node_idna_punycode_to_unicode(\n \n /// Converts a domain to ASCII as per the IDNA spec\n /// (specifically UTS #46)\n+///\n+/// Returns an empty string if the domain is invalid, matching Node.js behavior\n #[op2]\n #[string]\n-pub ... | 2025-12-03T14:57:53 |
vuejs/vue | 46b3bcd707addd0e503cc44bf7914b0281780d5c | 9931b715cdd163605e858adf8c42778cd3e34e71 | fix object dot notation bug (fix #4185) (#4188)
* fix object dot notation bug
* add test case | [
{
"path": "src/compiler/helpers.js",
"patch": "@@ -116,7 +116,7 @@ export function parseModel (val: string): Object {\n len = str.length\n index = expressionPos = expressionEndPos = 0\n \n- if (val.indexOf('[') < 0) {\n+ if (val.indexOf('[') < 0 || val.lastIndexOf(']') < len - 1) {\n return {\n ... | 2016-11-13T18:15:08 |
ollama/ollama | 7c6ea2a966bd7702922c81b05c7ea05bbcee67ba | c161aef5f970bc759c6241ec3d34e85324416190 | fix dangling """ | [
{
"path": "docs/modelfile.md",
"patch": "@@ -32,8 +32,6 @@ PARAMETER num_ctx 4096\n \n # Overriding the system prompt\n SYSTEM You are Mario from super mario bros, acting as an assistant.\n-\n-\"\"\"\n ```\n \n To use this:",
"additions": 0,
"deletions": 2,
"language": "Markdown"
}
] | 2023-07-20T06:10:32 |
ggml-org/llama.cpp | 304ac5693d1e2124f83e0584bc5eea6311d3d3b4 | 6c88ad8fa741a2182a2dcf8c5353ae4b5c66e656 | Vulkan iGPU device selection overhaul and PCI ID API support (#15947)
* vulkan: implement ggml igpu device type, implement pci id support
* fix compiler warning
* prevent printf overflow warning | [
{
"path": "ggml/src/ggml-vulkan/ggml-vulkan.cpp",
"patch": "@@ -4497,7 +4497,7 @@ static void ggml_vk_instance_init() {\n new_driver.pNext = &new_id;\n devices[i].getProperties2(&new_props);\n \n- if (new_props.properties.deviceType == vk::PhysicalDeviceType::eDiscreteGpu)... | 2025-09-12T11:24:21 |
huggingface/transformers | d848a3953aeb7debdbfbc6bfbf788c528cd40e46 | e4913bdf505e9184ac9647d13172b9ce1b38fd9a | Remove all instances of `is_safetensors_available` (#41233)
* safetensors is a core dep
* fix
* ok
* simplify branching
* keep it for now
---------
Co-authored-by: Cyril Vallez <cyril.vallez@gmail.com> | [
{
"path": "src/transformers/commands/env.py",
"patch": "@@ -14,7 +14,6 @@\n \n \n import contextlib\n-import importlib.util\n import io\n import os\n import platform\n@@ -26,7 +25,6 @@\n from ..integrations.deepspeed import is_deepspeed_available\n from ..utils import (\n is_accelerate_available,\n- ... | 2025-10-01T13:57:28 |
denoland/deno | 54412e4e8fd1fe3dadb9dfb9276a9dfccca90fd1 | 62e63ac9cacd6f55d5999c17814ef2d842213b5f | test: disable parallel/test-fs-promises-file-handle-readFile.js (#31485)
This test needs some more work, I've seen it crash multiple times in
various PRs over the last 24h. | [
{
"path": "tests/node_compat/config.toml",
"patch": "@@ -463,7 +463,8 @@\n \"parallel/test-fs-open-numeric-flags.js\" = {}\n \"parallel/test-fs-open.js\" = {}\n \"parallel/test-fs-promises-exists.js\" = {}\n-\"parallel/test-fs-promises-file-handle-readFile.js\" = { flaky = true }\n+# TODO(bartlomieju): cras... | 2025-12-03T14:46:43 |
vuejs/vue | 4f84aefe43d0943ba5d83e6bca876cd89cb6763a | 69eab86813b6d2cb1c8a7542fee6ccfe776988f5 | support v-model in weex (#4178)
* [wip] supported v-model in Weex
* fixed v-model in weex
* added test cases for v-model in weex
* used strToRegExp for all test cases in weex
* fixed eslint
* fixed parseModel test case | [
{
"path": "src/compiler/helpers.js",
"patch": "@@ -94,3 +94,87 @@ export function getAndRemoveAttr (el: ASTElement, name: string): ?string {\n }\n return val\n }\n+\n+let len, str, chr, index, expressionPos, expressionEndPos\n+\n+/**\n+ * parse directive model to do the array update transform. a[idx] = ... | 2016-11-12T16:33:51 |
Subsets and Splits
Assembly Language GitHub Issues
Retrieves a sample of assembly-language related commits with their details, but doesn't provide meaningful analysis or patterns beyond basic filtering.
Swift Compiler Issues Analysis
Retrieves all training data for the Swift programming language repository, providing basic filtering but offering limited analytical insight beyond identifying relevant code examples.