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 |
|---|---|---|---|---|---|
huggingface/transformers | cac0a28c83cf87b7a05495de3177099c635ba852 | 2a61590a479d3b1f77059f75caee7cc22760019d | Add afmoe model (#42168)
* Add AFMoE model support
* Address review feedback for AFMoE implementation
* Add flex attention support to AFMoE model
* Fix expert_bias routing in AFMoE
* Remove test-results directory
* Address PR review feedback for AFMoE model
* fix(afmoe): ensure RMSNorm output dtype matches input... | [
{
"path": ".gitignore",
"patch": "@@ -175,3 +175,4 @@ tags\n \n # Cursor IDE files\n .cursor/\n+test-results/",
"additions": 1,
"deletions": 0,
"language": "Unknown"
},
{
"path": "docs/source/en/_toctree.yml",
"patch": "@@ -384,6 +384,8 @@\n title: Main Classes\n - sections:\n ... | 2025-11-29T11:20:04 |
ggml-org/llama.cpp | a8c7f33d792c7ba8ecb889752f80945373e2ddfc | b7f5f46e03edbe73abb0784e27faa20efb8a42d5 | ci : change the cann version and the container pull method (#17953)
fix error format
Update build.yml
Remove unnecessary zip files
fix
update | [
{
"path": ".devops/cann.Dockerfile",
"patch": "@@ -4,7 +4,7 @@\n \n # Define the CANN base image for easier version updates later\n ARG CHIP_TYPE=910b\n-ARG CANN_BASE_IMAGE=quay.io/ascend/cann:8.3.rc1.alpha001-${CHIP_TYPE}-openeuler22.03-py3.11\n+ARG CANN_BASE_IMAGE=quay.io/ascend/cann:8.3.rc2-${CHIP_TYPE}-... | 2025-12-12T19:43:00 |
vuejs/vue | ebcef58645af1582ca3c8a19ec26967946970301 | f9f74231a2e064861856396f57dc980a6de376e9 | fix: fix v-for alias deconstruct regression
fix #7096 | [
{
"path": "src/compiler/error-detector.js",
"patch": "@@ -80,7 +80,7 @@ function checkIdentifier (\n ) {\n if (typeof ident === 'string') {\n try {\n- new Function(`var ${ident}`)\n+ new Function(`var ${ident}=_`)\n } catch (e) {\n errors.push(`invalid ${type} \"${ident}\" in expre... | 2017-11-21T14:06:19 |
denoland/deno | 0c8469433a80aa3b2513c9c462c55b0b6804319c | f1d7d0c2a75e1754920048f65a2c1526ff6d5ddf | fix(ext/node): implement `node:v8` heap APIs (#32483)
Implements `v8.getHeapSpaceStatistics()`,
`v8.getHeapCodeStatistics()`, `v8.getHeapSnapshot()`, and
`v8.writeHeapSnapshot()` from the `node:v8` module
Enabled node compat tests: `test-heapdump.js`,
`test-v8-getheapsnapshot-twice.js`, `test-v8-stats.js`
---------
... | [
{
"path": "Cargo.lock",
"patch": "@@ -10864,9 +10864,9 @@ dependencies = [\n \n [[package]]\n name = \"v8\"\n-version = \"146.1.0\"\n+version = \"146.3.0\"\n source = \"registry+https://github.com/rust-lang/crates.io-index\"\n-checksum = \"70bb6bcbb6a406fd720dad66d85337bd401d8276c7276f6a86164af6e1c57118\"\n... | 2026-03-05T14:20:13 |
ggml-org/llama.cpp | 482211438dd671224a7f176b7480b4ded424212c | 7bed317f5351eba037c2e0aa3dce617e277be1c4 | CUDA: fix overflow in MMA kernel without stream-k (#17939) | [
{
"path": "ggml/src/ggml-cuda/fattn-common.cuh",
"patch": "@@ -642,8 +642,8 @@ static __global__ void flash_attn_stream_k_fixup(\n const int iter_k = (ne11 + (nbatch_fa - 1)) / nbatch_fa;\n const int iter_j = (ne01 + (ncols1 - 1)) / ncols1;\n \n- const int kbc0 = (bidx0 + 0)*(iter_k*iter_... | 2025-12-12T16:43:58 |
vuejs/vue | d891cd1761df22e1e0b1953c6ed7947fdb79d915 | c264335fbd1b1d838e3c1085b7d6dcd1c752aa43 | fix: improve error detector v-for identifier check
close #6971 | [
{
"path": "src/compiler/error-detector.js",
"patch": "@@ -15,9 +15,6 @@ const unaryOperatorsRE = new RegExp('\\\\b' + (\n 'delete,typeof,void'\n ).split(',').join('\\\\s*\\\\([^\\\\)]*\\\\)|\\\\b') + '\\\\s*\\\\([^\\\\)]*\\\\)')\n \n-// check valid identifier for v-for\n-const identRE = /[A-Za-z_$][\\w$]*... | 2017-11-20T19:12:10 |
huggingface/transformers | 2a61590a479d3b1f77059f75caee7cc22760019d | a4a7008ee046b0ff6e38cd389080e1a3982b994f | Fix broken models due to modeling/weight mapping (#42468)
* lfm2
* fix
* config
* fixes
* fix prefixes
* finally
* copies
* oupsi
* rework key renaming
* fixes
* fix
* fix
* fix
* fix
* fix
* fix lfm2
* fix
* fix more bad mappings | [
{
"path": "src/transformers/conversion_mapping.py",
"patch": "@@ -70,11 +70,56 @@ def _build_checkpoint_conversion_mapping():\n operations=[MergeModulelist(dim=0), Concatenate(dim=1)],\n ),\n WeightConverter(\n- source_patterns=[\"mlp.experts.*.down_pro... | 2025-11-28T20:38:14 |
ollama/ollama | 531324a9bee6d1cd9330daa021388cccc6e2932f | a039e383cd133872c1b0599d2e960405a3e7d491 | feat: add OLLAMA_DEBUG in ollama server help message (#3461)
Co-authored-by: Pier Francesco Contino <pfcontino@gmail.com> | [
{
"path": "cmd/cmd.go",
"patch": "@@ -976,6 +976,7 @@ Environment Variables:\n OLLAMA_ORIGINS A comma separated list of allowed origins.\n OLLAMA_MODELS The path to the models directory (default is \"~/.ollama/models\")\n OLLAMA_KEEP_ALIVE The duration that models stay loaded in mem... | 2024-04-03T01:20:03 |
denoland/deno | f1d7d0c2a75e1754920048f65a2c1526ff6d5ddf | fb44def9e21e113ad2a679a44e54cd8f3a022665 | fix: prevent panic on dynamic import with non-string error name (#32498)
Fixes a panic where a dynamic import rejecting
with an Error whose `name` property is a non-string type (Number,
Object, Symbol) would panic and terminate the Deno process.
---------
Co-authored-by: kaju <kajukitli@users.noreply.github.com> | [
{
"path": "libs/core/error.rs",
"patch": "@@ -996,10 +996,14 @@ impl JsError {\n let mut out = Vec::with_capacity(arr.length() as usize);\n \n for i in 0..arr.length() {\n- let key = arr.get_index(scope, i).unwrap();\n+ let Some(key) = arr.get_index(scope, i) else {\n+ ... | 2026-03-05T14:16:36 |
ggml-org/llama.cpp | 7bed317f5351eba037c2e0aa3dce617e277be1c4 | dcb7d17758a95bada4383869c8ce66ca60ac3efc | models : fix the attn_factor for mistral3 graphs + improve consistency (#17945)
* models : fix the attn_factor for mistral3 graphs
* cont : rework attn_factor correction logic
* cont : make deepseek2 consistent
* cont : add TODO
* cont : special-case DSv2
* cont : revert Mistral 3 Large changes
* cont : fix DS2 ... | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -7286,6 +7286,10 @@ def set_gguf_parameters(self):\n self.gguf_writer.add_rope_scaling_type(gguf.RopeScalingType.YARN)\n self.gguf_writer.add_rope_scaling_factor(rope_scaling[\"factor\"])\n self.gguf_writer.add_rope_scaling_... | 2025-12-12T15:12:40 |
vuejs/vue | 4c22d1d17ffd3a9340c3b17443c7989d04ab14c5 | 063acb79ebc02344ab277196d4aea0577b113926 | fix: allow traversing reactive objects which are sealed (#7080) | [
{
"path": "src/core/observer/traverse.js",
"patch": "@@ -18,7 +18,7 @@ export function traverse (val: any) {\n function _traverse (val: any, seen: SimpleSet) {\n let i, keys\n const isA = Array.isArray(val)\n- if ((!isA && !isObject(val)) || !Object.isExtensible(val)) {\n+ if ((!isA && !isObject(val))... | 2017-11-20T15:38:39 |
huggingface/transformers | a4a7008ee046b0ff6e38cd389080e1a3982b994f | 6587d77ffcf647b848a71c32ce9cad3496e341b2 | [`Docs`] Fix build on other langs (#42485)
* fix
* albert missed ouch | [
{
"path": "docs/source/ja/model_doc/albert.md",
"patch": "@@ -60,9 +60,7 @@ ALBERTモデルは、「[ALBERT: A Lite BERT for Self-supervised Learning of Lan\n ## AlbertTokenizer\n \n [[autodoc]] AlbertTokenizer\n- - build_inputs_with_special_tokens\n - get_special_tokens_mask\n- - create_token_type_ids_from_s... | 2025-11-28T18:52:13 |
ggml-org/llama.cpp | dcb7d17758a95bada4383869c8ce66ca60ac3efc | 51604435e8efffb4f2e1991ddd2bbc01774dd0fc | cann : fix ops broken by circular padding guard (#17825) | [
{
"path": "ggml/src/ggml-cann/ggml-cann.cpp",
"patch": "@@ -2548,6 +2548,7 @@ static bool ggml_backend_cann_supports_op(ggml_backend_dev_t dev, const ggml_ten\n case GGML_OP_ARGSORT:\n case GGML_OP_ACC:\n case GGML_OP_GROUP_NORM:\n+ return true;\n case GGML_OP_PAD:... | 2025-12-12T14:49:27 |
denoland/deno | ffdbefb7331321dd8bc21272ea1631165e310b73 | 576871afb297c36af300bcb6009bb59a53d8ace4 | fix(deploy): correctly handle version lookup (#32489) | [
{
"path": "cli/tools/deploy.rs",
"patch": "@@ -47,7 +47,7 @@ pub async fn deploy(\n let file = factory\n .file_fetcher()?\n .fetch_bypass_permissions(\n- ®istry_url.join(\"@deploy/deno/meta.json\").unwrap(),\n+ ®istry_url.join(\"@deno/deploy/meta.json\").unwrap(... | 2026-03-05T11:12:30 |
vuejs/vue | 063acb79ebc02344ab277196d4aea0577b113926 | a23b913796a7d18e76185607f250655e18a390c8 | fix(ssr): fix bundle renderer require path on windows (#7085)
fixed windows os `vue-ssr-server-bundle.json` within module can't found on server rendering | [
{
"path": "src/server/bundle-renderer/create-bundle-runner.js",
"patch": "@@ -51,7 +51,7 @@ function compileModule (files, basedir, runInNewContext) {\n : script.runInNewContext(sandbox)\n const m = { exports: {}}\n const r = file => {\n- file = path.join('.', file)\n+ file = path.po... | 2017-11-20T15:33:53 |
ollama/ollama | 80163ebcb5f79661ae0c7094cc2983c87dab2e28 | a57818d93e251317d32df40d9bb8d7df8a4075bf | fix metal gpu | [
{
"path": "llm/server.go",
"patch": "@@ -85,7 +85,7 @@ func NewLlamaServer(model string, adapters, projectors []string, opts *api.Optio\n \tgraph := int64(ggml.KV().GQA()) * kv / 6\n \tusedMemory += graph\n \n-\tif usedMemory > availableMemory || slices.Contains(cpuOnlyFamilies, ggml.KV().Architecture()) {\... | 2024-04-02T23:06:45 |
ggml-org/llama.cpp | 51604435e8efffb4f2e1991ddd2bbc01774dd0fc | 17158965ac95b8bea2cbeff559f4cdba5d44c36f | ggml-cpu : fix RISC-V Q4_0 repack select and RVV feature reporting (#17951)
* ggml-cpu:fix RISC-V Q4_0 repack select and RVV feature reporting
Signed-off-by: Wang Yang <yangwang@iscas.ac.cn>
* using the name VLEN instead of CNT
* Update ggml/include/ggml-cpu.h
---------
Signed-off-by: Wang Yang <yangwang@iscas.ac... | [
{
"path": "ggml/include/ggml-cpu.h",
"patch": "@@ -99,6 +99,7 @@ extern \"C\" {\n GGML_BACKEND_API int ggml_cpu_has_sme (void);\n // other\n GGML_BACKEND_API int ggml_cpu_has_riscv_v (void);\n+ GGML_BACKEND_API int ggml_cpu_get_rvv_vlen (void); // risc-v vector length in bytes\n ... | 2025-12-12T14:26:03 |
huggingface/transformers | 6587d77ffcf647b848a71c32ce9cad3496e341b2 | 4abfd165e6175012a6509284b3ad482235eab1d7 | add rotary kernel support to Qwen3 model (#41147)
* add rotary kernel support to Qwen3 model
Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
* delete unnecessary import
Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
* adjust code
Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
* adjust code
Signed-... | [
{
"path": "src/transformers/integrations/__init__.py",
"patch": "@@ -72,6 +72,7 @@\n \"register_kernel_mapping\",\n \"replace_kernel_forward_from_hub\",\n \"use_kernel_forward_from_hub\",\n+ \"use_kernel_func_from_hub\",\n ],\n \"integration_utils\": [\n \"INTE... | 2025-11-28T17:24:34 |
denoland/deno | 576871afb297c36af300bcb6009bb59a53d8ace4 | cc53b5ff396a1d40f3f88712a506395356fb57bc | ci: maybe fix spec test timeout (#32487) | [
{
"path": "tests/util/lib/builders.rs",
"patch": "@@ -385,9 +385,16 @@ impl TestContext {\n \n fn kill_process(pid: u32) {\n #[cfg(unix)]\n- // SAFETY: We're sending SIGKILL to a process we spawned.\n- unsafe {\n- libc::kill(pid as i32, libc::SIGKILL);\n+ {\n+ // kill the process tree by first ki... | 2026-03-05T10:47:07 |
vuejs/vue | a23b913796a7d18e76185607f250655e18a390c8 | b2781205b0271a9b59afe0c816b41c6dbe85f0a3 | fix: fix <keep-alive> include/exclude logic for anonymous components | [
{
"path": "src/core/components/keep-alive.js",
"patch": "@@ -87,10 +87,13 @@ export default {\n if (componentOptions) {\n // check pattern\n const name: ?string = getComponentName(componentOptions)\n- if (!name || (\n- (this.exclude && matches(this.exclude, name)) ||\n- (t... | 2017-11-20T15:04:06 |
ollama/ollama | 841adda157209d6498d65a3d969b96f2fdd64300 | 0035e31af87c6cc62cde5d6cd0f22d36131ff62e | Fix windows lint CI flakiness | [
{
"path": ".github/workflows/test.yaml",
"patch": "@@ -260,7 +260,9 @@ jobs:\n mkdir -p llm/build/windows/$ARCH/stub/stub/bin/\n touch llm/build/windows/$ARCH/stub/stub/bin/stub.dll\n if: ${{ startsWith(matrix.os, 'windows-') }}\n- - uses: golangci/golangci-lint-action@v3\n+... | 2024-04-02T19:22:16 |
ggml-org/llama.cpp | 12280ae9058600b0b14335fe59848e601e10eebc | 54a0fee4b73e9023c6faef9b2cfe4963edb6776c | webui: Fix parsing non-LaTeX occurrencies of `\(` or `\)` (#17810)
* fix: Improve latex protection logic to prevent turning non-latex `\(` into `$`
* chore: update webui build output | [
{
"path": "tools/server/webui/src/lib/utils/latex-protection.test.ts",
"patch": "@@ -303,6 +303,27 @@ $$\\n\\\\pi_n(\\\\mathbb{S}^3) = \\\\begin{cases}\n \t\texpect(output).toBe(input); // Code blocks prevent misinterpretation\n \t});\n \n+\ttest('preserves backslash parentheses in code blocks (GitHub issue... | 2025-12-12T14:13:36 |
huggingface/transformers | 4abfd165e6175012a6509284b3ad482235eab1d7 | c79ac0717f4484c1e4af156c36315282a5bc5db0 | Fix minor bugs in ConversionOps and WeightConverter (#42479)
* style
* fix
* fix
* Revert "fix"
This reverts commit b7fc6aa2688f4a8ca14047411c78f25822da7499.
* remove reverse for now as not needed
---------
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com> | [
{
"path": "src/transformers/core_model_loading.py",
"patch": "@@ -105,7 +105,10 @@ class ConversionOps:\n \"\"\"Base class for weight conversion operations.\"\"\"\n \n def __repr__(self):\n- return f\"{self.__class__.__name__}(dim={self.dim})\"\n+ if hasattr(self, \"dim\"):\n+ ... | 2025-11-28T16:46:57 |
vuejs/vue | 59dea374037ec2e6b1f5570a30774f2de0a44adc | 06f121b4735e8c70306c744419287592af295eac | fix: fix v-model :value warning on custom component
fix #7084 | [
{
"path": "src/platforms/web/compiler/directives/model.js",
"patch": "@@ -21,7 +21,6 @@ export default function model (\n const modifiers = dir.modifiers\n const tag = el.tag\n const type = el.attrsMap.type\n- const attrsMap = el.attrsMap\n \n if (process.env.NODE_ENV !== 'production') {\n // i... | 2017-11-18T19:41:22 |
denoland/deno | cc53b5ff396a1d40f3f88712a506395356fb57bc | 9d90318a53cce7657d31e28814f5ab5d68f00d68 | fix(test): restart all tests on env file change (#32461)
Closes https://github.com/denoland/deno/issues/32222 | [
{
"path": "cli/args/mod.rs",
"patch": "@@ -640,8 +640,16 @@ impl CliOptions {\n self.flags.no_legacy_abort()\n }\n \n- pub fn env_file_name(&self) -> Option<&Vec<String>> {\n- self.flags.env_file.as_ref()\n+ pub fn env_file_paths(\n+ &self,\n+ ) -> impl DoubleEndedIterator<Item = PathBuf> + '... | 2026-03-05T10:46:42 |
ggml-org/llama.cpp | dada4c846df5e2bea1800e70b56ceacac7cfb54b | b8ee22cfde0eac9e20ce7963c02138a74c9471f7 | model-conversion : remove max diff check in compare-logits [no ci] (#17954)
This commit removes the maximum difference check from the
compare-logits.py which would stop early if the difference between
the logits exceeded a threshold.
The motivation for removing this is that it can be useful to be able to
get the comp... | [
{
"path": "examples/model-conversion/scripts/causal/compare-logits.py",
"patch": "@@ -32,10 +32,6 @@ def quick_logits_check(pytorch_file, llamacpp_file):\n print(f\"Top 10 llama.cpp logits: {llamacpp_logits[llamacpp_top10]}\")\n print(f\"Max absolute difference: {max_diff:.4f}\")\n \n- if max_dif... | 2025-12-12T12:25:16 |
huggingface/transformers | c79ac0717f4484c1e4af156c36315282a5bc5db0 | f8e29b5a1b9d5f69c187b68351cc12988b3786cb | Fix gpt2 tokenizer `add_prefix_space` default value (#42481)
fix tok | [
{
"path": "src/transformers/models/gpt2/tokenization_gpt2.py",
"patch": "@@ -101,7 +101,7 @@ def __init__(\n bos_token=\"<|endoftext|>\",\n eos_token=\"<|endoftext|>\",\n pad_token=None,\n- add_prefix_space=True,\n+ add_prefix_space=False,\n add_bos_token=False,... | 2025-11-28T15:37:18 |
vuejs/vue | 4d8226fb2c84fa2e13a2d8a86dea8a9a5c6ea95f | 6d6b3739e132723915bc2209663db1b825307865 | fix: keep-alive should not cache anonymous components
This only happens if the component is returned by a intermediate
functional or abstract component, e.g. <router-view>. Fix #6938. | [
{
"path": "src/core/components/keep-alive.js",
"patch": "@@ -87,7 +87,7 @@ export default {\n if (componentOptions) {\n // check pattern\n const name: ?string = getComponentName(componentOptions)\n- if (name && (\n+ if (!name || (\n (this.exclude && matches(this.exclude, na... | 2017-11-17T15:30:51 |
denoland/deno | 9d90318a53cce7657d31e28814f5ab5d68f00d68 | 888568c66267ed957423937764e22081d526cc08 | fix(ext/node): use sync writes for TTY stdout/stderr WriteStream (#32488)
## Summary
- After #32343 switched `process.stdout`/`process.stderr` to
`tty.WriteStream` (extending `net.Socket`), writes became asynchronous
through the `LibuvStreamWrap` handle
- Since `console.log` in Deno writes synchronously via
`Deno.cor... | [
{
"path": "ext/node/polyfills/internal/tty.js",
"patch": "@@ -10,6 +10,7 @@ const {\n Error,\n ObjectEntries,\n ObjectPrototypeHasOwnProperty,\n+ ObjectPrototypeIsPrototypeOf,\n ObjectValues,\n RegExpPrototypeExec,\n SafeMap,\n@@ -22,6 +23,8 @@ const {\n SetPrototypeGetSize,\n StringPrototy... | 2026-03-05T10:44:46 |
huggingface/transformers | f8e29b5a1b9d5f69c187b68351cc12988b3786cb | 7b5f1db802e14d1f0a45d9854c4772bae46aa594 | [quantization] make torchao tests slow (#42482)
* fix
* fix | [
{
"path": "tests/quantization/torchao_integration/test_torchao.py",
"patch": "@@ -29,6 +29,7 @@\n require_torch_multi_accelerator,\n require_torchao,\n require_torchao_version_greater_or_equal,\n+ slow,\n torch_device,\n )\n from transformers.utils import is_torch_available, is_torchao_av... | 2025-11-28T15:27:44 |
vuejs/vue | 6d6b3739e132723915bc2209663db1b825307865 | f5cd29e1d8197613c4dfb4013b240784c3b64e43 | fix: should warn unknown components inside <keep-alive> | [
{
"path": "src/core/components/keep-alive.js",
"patch": "@@ -81,7 +81,8 @@ export default {\n },\n \n render () {\n- const vnode: VNode = getFirstComponentChild(this.$slots.default)\n+ const slot = this.$slots.default\n+ const vnode: VNode = getFirstComponentChild(slot)\n const componentOpt... | 2017-11-17T15:23:23 |
denoland/deno | d143d7c16110debcf069fbd2b53e9287e1a495d3 | be569cf522240bf1b0ec4ebc906ca69415723891 | fix(ext/node): deep `assert` compatibility (#32434) | [
{
"path": "ext/crypto/00_crypto.js",
"patch": "@@ -90,6 +90,7 @@ const {\n import * as webidl from \"ext:deno_webidl/00_webidl.js\";\n import { createFilteredInspectProxy } from \"ext:deno_web/01_console.js\";\n import { DOMException } from \"ext:deno_web/01_dom_exception.js\";\n+import { kKeyObject } from ... | 2026-03-04T23:08:56 |
ggml-org/llama.cpp | a81a569577cc38b32558958b048228150be63eae | 53ecd4fdb923dcb53d311ed42798cae7198aa742 | Add a search field on model selector / improve mobile display (#17765)
* webui: add search field to model selector and fixes mobile viewport overflow
* webui: simplify model search style and code
* refacor: Search Input component & consistent UI for Models Selector search
* feat: Use Popover component + improve int... | [
{
"path": "tools/server/webui/package-lock.json",
"patch": "@@ -41,7 +41,7 @@\n \t\t\t\t\"@tailwindcss/vite\": \"^4.0.0\",\n \t\t\t\t\"@types/node\": \"^22\",\n \t\t\t\t\"@vitest/browser\": \"^3.2.3\",\n-\t\t\t\t\"bits-ui\": \"^2.8.11\",\n+\t\t\t\t\"bits-ui\": \"^2.14.4\",\n \t\t\t\t\"clsx\": \"^2.1.1\",\n ... | 2025-12-11T17:21:21 |
huggingface/transformers | 230c3234cb8720fc9bf467e55992c02aaa1db109 | 078ff685d37a8efb4966bd3da322eef92bf875c0 | Fix doc builds (#42478)
fix doc builds | [
{
"path": "src/transformers/models/llama/tokenization_llama.py",
"patch": "@@ -77,15 +77,6 @@ class LlamaTokenizer(TokenizersBackend):\n Whether or not to add an `eos_token` at the end of sequences.\n use_default_system_prompt (`bool`, *optional*, defaults to `False`):\n Whet... | 2025-11-28T14:44:23 |
vuejs/vue | f5cd29e1d8197613c4dfb4013b240784c3b64e43 | 1e146037fa4280b502d0edf95936bc67e87fd339 | fix: init _staticTrees to avoid runtime reference warning
fix #7075 | [
{
"path": "flow/component.js",
"patch": "@@ -66,6 +66,7 @@ declare interface Component {\n _isDestroyed: boolean;\n _isBeingDestroyed: boolean;\n _vnode: ?VNode; // self root node\n+ _staticTrees: ?Array<VNode>; // v-once cached trees\n _hasHookEvent: boolean;\n _provided: ?Object;\n ",
"addi... | 2017-11-17T14:45:03 |
denoland/deno | be569cf522240bf1b0ec4ebc906ca69415723891 | 1e7a295613519a1f9aa87c0a3f0566039df944eb | fix(core): uv_compat cleanup and fixes (#32458)
- split uv_compat into multiple files
- move logic out of run_io into helpers
- avoid holding references across callbacks to prevent violating
aliasing
- add a test suite, this goes through the same JsRuntime event loop
infra as deno
- fix shutdown not waiting for write ... | [
{
"path": "libs/core/runtime/jsruntime.rs",
"patch": "@@ -2164,7 +2164,18 @@ impl JsRuntime {\n Self::dispatch_rejections(scope, context_state, exception_state)?;\n scope.perform_microtask_checkpoint();\n \n- // ===== Phase 3: I/O =====\n+ // ===== Phase 3: Idle / Prepare =====\n+ // In lib... | 2026-03-04T17:36:20 |
ggml-org/llama.cpp | 53ecd4fdb923dcb53d311ed42798cae7198aa742 | c6f6e4f96a7f7bce49f5c21d19ee69fb8b72f84d | SOLVE_TRI extension to more dimensions (#17793)
* Extended TRI
* Fix whitespace
* chore: update webui build output
* Just use cuBLAS for everything...
* Merge both versions
* Remove incorrect imports causing failures for CI
* Still failing... remove all direct cublas imports and rely on common imports from "comm... | [
{
"path": "ggml/src/ggml-cuda/ggml-cuda.cu",
"patch": "@@ -4630,9 +4630,9 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g\n case GGML_OP_CUMSUM:\n case GGML_OP_TRI:\n case GGML_OP_DIAG:\n- return true;\n case GGML_OP_SOLVE_TRI:\n... | 2025-12-11T16:20:43 |
vuejs/vue | 1e146037fa4280b502d0edf95936bc67e87fd339 | a2cd14c765cfb1f881f34911d4fd0540feb09cce | feat: warn if both v-model and v-bind:value used on same element (#7056)
* test($compile): warn if v-model and :value used on same text input
#7048
* test($compile): make v-model and v-bind:value warning apply to all but exceptions
#7048
* test($compile): move v-model/:value conflict warner to model.js
... | [
{
"path": "src/platforms/web/compiler/directives/model.js",
"patch": "@@ -21,6 +21,7 @@ export default function model (\n const modifiers = dir.modifiers\n const tag = el.tag\n const type = el.attrsMap.type\n+ const attrsMap = el.attrsMap\n \n if (process.env.NODE_ENV !== 'production') {\n // i... | 2017-11-16T16:24:23 |
huggingface/transformers | 078ff685d37a8efb4966bd3da322eef92bf875c0 | 554fb40e8275fede4df538c6825f777d9b18dcd5 | 🚨 Move `rotary_partial_emb` to RopeParams and delete unnecessary code 🔪 (#42255)
* tmp
* batch push
* maybe better pop and break, and we'll have one theta per config in the rope dict
* update a few models?
* fix tests that are easu first
* dont overwrite if already present!!!
* partial rotary factor
* more f... | [
{
"path": "MIGRATION_GUIDE_V5.md",
"patch": "@@ -328,7 +328,7 @@ model_4bit = AutoModelForCausalLM.from_pretrained(\n \n - Methods to init a nested config such as `from_xxx_config` are deleted. Configs can be init from the `__init__` method in the same way. See [#41314](https://github.com/huggingface/transf... | 2025-11-28T12:45:40 |
ollama/ollama | 1ec0df1069341cf09417401c37474b92999d3c12 | 011bb6735187b6c5898ab7504ffe473057b2eed1 | fix generate output | [
{
"path": ".github/workflows/test.yaml",
"patch": "@@ -12,6 +12,7 @@ jobs:\n changes:\n runs-on: ubuntu-latest\n outputs:\n+ GENERATE: ${{ steps.changes.outputs.GENERATE }}\n GENERATE_CUDA: ${{ steps.changes.outputs.GENERATE_CUDA }}\n GENERATE_ROCM: ${{ steps.changes.outputs.GENER... | 2024-04-01T20:47:34 |
denoland/deno | 92ed54bba86a5e1fc6ee3c4d7eaa668a00978edd | 8426078dae05b031c13f37495dd78acb105c5583 | Reland refactor(core): move nextTick/immediate queues into core, replace ops with shared buffers (#32440) (#32465)
## Summary
- The nextTick rewrite (#32440) moved unhandled rejection processing
into `processTicksAndRejections` (JS-side `processPromiseRejections`),
but `rejectionhandled` events were still only dispatc... | [
{
"path": "ext/node/lib.rs",
"patch": "@@ -493,7 +493,6 @@ deno_core::extension!(deno_node,\n \"internal/errors/error_source.ts\",\n \"internal/event_target.mjs\",\n \"internal/events/abort_listener.mjs\",\n- \"internal/fixed_queue.ts\",\n \"internal/fs/streams.mjs\",\n \"internal/fs/... | 2026-03-04T16:46:52 |
vuejs/vue | 5db86b4e94857fdde3ae6b71e24da637bc116baa | 6b799190a57b73c4b3d6c31f4009cda74a1a6184 | fix(ssr): ensure hydrated class & style bindings are reactive
fix #7063 | [
{
"path": "src/core/observer/traverse.js",
"patch": "@@ -0,0 +1,39 @@\n+/* @flow */\n+\n+import { _Set as Set, isObject } from '../util/index'\n+import type { SimpleSet } from '../util/index'\n+\n+const seenObjects = new Set()\n+\n+/**\n+ * Recursively traverse an object to evoke all converted\n+ * getters,... | 2017-11-16T16:19:52 |
ollama/ollama | e1f1c374ea3033d55e4975fbb60a5873f716b8ba | 06a1508bfe456e82ba053ea554264e140c5057b5 | Community Integration: ChatOllama (#3400)
* Community Integration: ChatOllama
* fixed typo | [
{
"path": "README.md",
"patch": "@@ -289,6 +289,7 @@ See the [API documentation](./docs/api.md) for all endpoints.\n - [AnythingLLM (Docker + MacOs/Windows/Linux native app)](https://github.com/Mintplex-Labs/anything-llm)\n - [Ollama Basic Chat: Uses HyperDiv Reactive UI](https://github.com/rapidarchitect/o... | 2024-03-31T02:46:50 |
ggml-org/llama.cpp | c6f6e4f96a7f7bce49f5c21d19ee69fb8b72f84d | d9f8f60618a1df2797cb7df4ad1272f71d6bd7b2 | ggml-alloc : fix reuse-parent logic for misaligned sizes (#17884) | [
{
"path": "ggml/src/ggml-alloc.c",
"patch": "@@ -312,16 +312,9 @@ static struct buffer_address ggml_dyn_tallocr_alloc(struct ggml_dyn_tallocr * al\n }\n \n // this is a very naive implementation, but for our case the number of free blocks should be very small\n-static void ggml_dyn_tallocr_free_tensor(struc... | 2025-12-11T12:30:10 |
denoland/deno | 8426078dae05b031c13f37495dd78acb105c5583 | 9f02e0019c3741ebf295b5a9e062b0b66a6ef18c | ci: fix deno_core changes check in shallow clones (#32464)
## Summary
- The previous shell one-liner used `git diff` against the PR base SHA,
which often isn't available in the shallow clone (`fetch-depth: 5`).
When the base SHA is missing, `git diff` fails silently, `grep` gets no
input, and deno_core tests are incor... | [
{
"path": ".github/workflows/ci.generate.ts",
"patch": "@@ -480,11 +480,9 @@ const preBuildCheckStep = step({\n const denoCoreChangesCheckStep = step({\n id: \"deno_core_changes\",\n run: [\n- `DENO_CORE_CHANGED=$(git diff --name-only \\${{ github.event.pull_request.base.sha }}..HEAD | grep -qE '^(${... | 2026-03-04T16:43:58 |
vuejs/vue | 6b799190a57b73c4b3d6c31f4009cda74a1a6184 | 45d7ba842917a075d6cb2563c78210e3b9210a58 | test: add test case for #6687 + fix lingering comment node | [
{
"path": "src/platforms/web/runtime/modules/transition.js",
"patch": "@@ -181,12 +181,12 @@ export function leave (vnode: VNodeWithData, rm: Function) {\n }\n \n const data = resolveTransition(vnode.data.transition)\n- if (isUndef(data)) {\n+ if (isUndef(data) || el.nodeType !== 1) {\n return rm(... | 2017-11-16T15:09:15 |
huggingface/transformers | b54f79073c5ec482df13304231edddf176c19cb0 | 6db4332171df2b4099c44c7a5c01258b91f7394a | replace source_keys and target_keys (#42471)
fix | [
{
"path": "src/transformers/quantizers/quantizer_bnb_4bit.py",
"patch": "@@ -306,7 +306,7 @@ def get_weight_conversions(self):\n if self.pre_quantized:\n return [\n WeightConverter(\n- source_keys=[\n+ source_patterns=[\n ... | 2025-11-28T12:41:22 |
ggml-org/llama.cpp | d9f8f60618a1df2797cb7df4ad1272f71d6bd7b2 | e4ae38331702aeb43b6ecc3f912d626171c9862a | batch : fix sequence id ownership (#17915)
* batch : fix sequence id ownage
* cont : reduce allocations | [
{
"path": "src/llama-batch.cpp",
"patch": "@@ -695,6 +695,8 @@ llama_ubatch llama_batch_allocr::ubatch_add(const std::vector<int32_t> & idxs, u\n udata->seq_idx .resize(LLAMA_MAX_SEQ, -1);\n udata->output .resize(n_tokens);\n \n+ udata->seq_id_data.reserve(n_tokens);\n+\n seq_set_t seq_s... | 2025-12-11T12:29:47 |
ollama/ollama | 06a1508bfe456e82ba053ea554264e140c5057b5 | 5a5efee46b2b633ea0e8d69311b57ab4c2d71abe | Update 90_bug_report.yml | [
{
"path": ".github/ISSUE_TEMPLATE/90_bug_report.yml",
"patch": "@@ -19,7 +19,7 @@ body:\n label: What did you expect to see?\n description: What did you expect to see/happen instead?\n validations:\n- required: true\n+ required: false\n - type: textarea\n id: steps\n attr... | 2024-03-29T14:11:17 |
huggingface/transformers | 6db4332171df2b4099c44c7a5c01258b91f7394a | c95d4affc1198cd9c95c41063d0b71a1c9c8e07c | [Trainer] use output.loss when using liger-kernel (#42444)
* use output.loss when using liger
Handle loss computation for models using Liger-kernel.
fixes #42414
* Clarify Liger-kernel loss computation in comments
* Both standard transformers and Liger models handle shift_labels correctly via **kwargs
* removed u... | [
{
"path": "src/transformers/trainer.py",
"patch": "@@ -3909,7 +3909,7 @@ def compute_loss(\n \n def _deepspeed_sp_compute_loss(self, model, inputs, return_outputs, pc):\n \"\"\"\n- How the loss is computed by Trainer under sequence parallelism with sp_backend==\"deepspeed\" and sp_size>1.... | 2025-11-28T11:00:33 |
vuejs/vue | 45d7ba842917a075d6cb2563c78210e3b9210a58 | a3246fd8545dbe66f37da16d7ff68597c0f025e6 | fix(transition): fix out-in transition getting stuck with v-if (#7023)
fix #6687 | [
{
"path": "src/platforms/web/runtime/components/transition.js",
"patch": "@@ -161,7 +161,9 @@ export default {\n oldChild &&\n oldChild.data &&\n !isSameChild(child, oldChild) &&\n- !isAsyncPlaceholder(oldChild)\n+ !isAsyncPlaceholder(oldChild) &&\n+ // #6687 component root ... | 2017-11-16T14:29:26 |
denoland/deno | 91e84aeffcf4e3cd00ec90c93cdef1a4810da3c4 | e78a7fa8d86848a759e6743537cf90013c7160b1 | refactor(core): move nextTick/immediate queues into core, replace ops with shared buffers (#32440)
## Summary
Major refactoring of the nextTick/immediate queue infrastructure, moving
it from Node polyfills into deno_core and replacing op-based
communication with shared memory buffers.
### Move nextTick queue into co... | [
{
"path": "ext/node/lib.rs",
"patch": "@@ -493,7 +493,6 @@ deno_core::extension!(deno_node,\n \"internal/errors/error_source.ts\",\n \"internal/event_target.mjs\",\n \"internal/events/abort_listener.mjs\",\n- \"internal/fixed_queue.ts\",\n \"internal/fs/streams.mjs\",\n \"internal/fs/... | 2026-03-04T15:23:24 |
ggml-org/llama.cpp | 34ce48d97a8cd5497ee418224da5bf422ed96673 | 45e350e3d3a663714d7d3d2397e14c2904534338 | ggml-hexagon: fix `rope` failure at `test-backend-ops` (#17565)
* fix test failure
* fix: correct scaling calculations in rope_cache_init
* fix: optimize element copying in rope_hex_f32 using memcpy
* fix: optimize loop boundaries in rope_hex_f32 for better performance
* feat: add profiling macros for performance ... | [
{
"path": "ggml/src/ggml-hexagon/htp/rope-ops.c",
"patch": "@@ -73,15 +73,15 @@ static float rope_yarn_ramp(const float low, const float high, const int i0) {\n return (1 - MIN(1, MAX(0, y)));\n }\n \n-static void rope_cache_init(const float theta_base,\n- float freq... | 2025-12-10T22:45:43 |
huggingface/transformers | 603eb77a12b9ba5a3ed1482e1dd92868599bdf91 | d08b98b965176ea9cf8c8e8b24995c955b7e2ec9 | Fix T5 tests: use generation_config for generation parameters (#42419)
* pass the generation parameters to generate()
* fix use_task_specific_params to separate model.config and model.generation_config params
* fix style
* some fixes
* remove redundant check
* update expectation for llama_7b_bf16 on rocm
* Updat... | [
{
"path": "tests/models/llama/test_modeling_llama.py",
"patch": "@@ -117,7 +117,7 @@ def test_model_7b_logits_bf16(self):\n (\"xpu\", 3): torch.tensor([[-6.5208, -4.1218, -4.9377, -3.2536, 0.8127, -2.9811, 1.2918, -3.3848]]),\n (\"cuda\", 7): torch.tensor([[-6.5061, -4.1147, -4.966... | 2025-11-28T08:32:22 |
denoland/deno | f85c0caafaa7f44c1b14d10babdfa4daaab8738f | 739f9ae0056dd4f3e6e1fec81ed75cc55c77b7f1 | fix(deploy): dont use jsr api for deploy subcommand (#32441) | [
{
"path": "cli/registry.rs",
"patch": "@@ -53,12 +53,6 @@ pub struct PublishingTask {\n pub error: Option<PublishingTaskError>,\n }\n \n-#[derive(serde::Deserialize)]\n-#[serde(rename_all = \"camelCase\")]\n-pub struct Package {\n- pub latest_version: Option<String>,\n-}\n-\n #[derive(serde::Deserialize)... | 2026-03-04T12:33:37 |
vuejs/vue | 13196b25b8a0a84b3936982177195d2e04f13f79 | df82aeb0bf7454ac99d403000a1ac993e8d8d4de | fix: clone slot nodes for render fn usage as well
fix #7041 | [
{
"path": "src/core/instance/render-helpers/resolve-slots.js",
"patch": "@@ -44,7 +44,7 @@ export function resolveSlots (\n }\n \n function isWhitespace (node: VNode): boolean {\n- return node.isComment || node.text === ' '\n+ return (node.isComment && !node.asyncFactory) || node.text === ' '\n }\n \n exp... | 2017-11-14T02:43:20 |
ggml-org/llama.cpp | 45e350e3d3a663714d7d3d2397e14c2904534338 | c6b2c9310cc53e5cd4f65bbb8d0cb498caf2ed1e | ci: fix riscv64-native build (#17916) | [
{
"path": ".github/workflows/build.yml",
"patch": "@@ -1770,7 +1770,7 @@ jobs:\n echo \"Fetch llama2c model\"\n wget https://huggingface.co/karpathy/tinyllamas/resolve/main/stories260K/stories260K.bin\n ./bin/llama-convert-llama2c-to-ggml --copy-vocab-from-model ./tok512.bin --... | 2025-12-10T22:24:31 |
denoland/deno | 739f9ae0056dd4f3e6e1fec81ed75cc55c77b7f1 | 7a5efce39115c24c652f1ffa74aca055352f3abd | fix(npm): don't panic when peer dep is not in snapshot (#32454)
Instead of panicking, create the missing node gracefully. | [
{
"path": "libs/npm/resolution/graph.rs",
"patch": "@@ -428,10 +428,10 @@ impl Graph {\n // package-b@1.0.0_package-c@1.0.0__package-b@1.0.0\n // ^ attempting to resolve this\n // In this case, we go up the ancestors to see if we can find a ... | 2026-03-04T11:51:56 |
huggingface/transformers | d08b98b965176ea9cf8c8e8b24995c955b7e2ec9 | 05c0e1d39082ee8b69064ed4ea9c239cd17405e9 | [loading/saving] Reverse all loading operations when saving (#42396)
* first shot
* default to reversing
* oupso
* oupsi 2
* oupsi 3
* fix renamed kwargs
* fix timm_wrapper
* remove fix_state_dict methods
* can do it all the time, with __init__ as well
* doc
* oupsi
* fix
* create helper
* fix annotation ... | [
{
"path": "src/transformers/conversion_mapping.py",
"patch": "@@ -13,7 +13,10 @@\n # See the License for the specific language governing permissions and\n # limitations under the License.\n \n+from __future__ import annotations\n+\n from copy import deepcopy\n+from typing import TYPE_CHECKING\n \n from .cor... | 2025-11-27T18:46:09 |
vuejs/vue | df82aeb0bf7454ac99d403000a1ac993e8d8d4de | daed1e73557d57df244ad8d46c9afff7208c9a2d | fix: should warn unknown components during hydration
fix #6998 | [
{
"path": "src/core/vdom/patch.js",
"patch": "@@ -103,7 +103,23 @@ export function createPatchFunction (backend) {\n }\n }\n \n- let inPre = 0\n+ function isUnknownElement (vnode, inVPre) {\n+ return (\n+ !inVPre &&\n+ !vnode.ns &&\n+ !(\n+ config.ignoredElements.length &&\n... | 2017-11-13T20:52:01 |
denoland/deno | 7a5efce39115c24c652f1ffa74aca055352f3abd | fce52fa2e9adcb236c4f01072e423601ec33b2f3 | fix(npm): regression where resolution was missing dependencies (#32453) | [
{
"path": "libs/npm/resolution/graph.rs",
"patch": "@@ -3225,7 +3225,9 @@ impl<'a, TNpmRegistryApi: NpmRegistryApi>\n .peer_dep_version_req\n .as_ref()\n .unwrap_or(&dep.version_req);\n- if versions.contains_key(effective_req) {\n+ if let Some(target_version) = ve... | 2026-03-04T11:48:20 |
ggml-org/llama.cpp | c6b2c9310cc53e5cd4f65bbb8d0cb498caf2ed1e | 34a6d86982b54314516fd40ef5110525247528b8 | mtmd: some small clean up (#17909)
* clip: add support for fused qkv in build_vit
* use bulid_ffn whenever possible
* fix internvl
* mtmd-cli: move image to beginning
* test script: support custom args | [
{
"path": "tools/mtmd/clip.cpp",
"patch": "@@ -595,11 +595,12 @@ struct clip_graph {\n cur = ggml_mul(ctx0, cur, model.mm_input_norm_w);\n cur = ggml_add(ctx0, cur, model.mm_input_norm_b);\n \n- cur = ggml_mul_mat(ctx0, model.mm_1_w, cur);\n- cur = ggml_add(ctx0... | 2025-12-10T21:20:06 |
vuejs/vue | daed1e73557d57df244ad8d46c9afff7208c9a2d | 7cf188e134fe7bfc9e8a16b763fb85b18eff1eac | fix: normlaize @click.right and @click.middle
fix #7020 | [
{
"path": "src/compiler/codegen/events.js",
"patch": "@@ -41,18 +41,7 @@ export function genHandlers (\n ): string {\n let res = isNative ? 'nativeOn:{' : 'on:{'\n for (const name in events) {\n- const handler = events[name]\n- // #5330: warn click.right, since right clicks do not actually fire cl... | 2017-11-13T17:55:56 |
ollama/ollama | 5255d0af8ad66f2919b3abcd2afcae7a95d2cdca | 461ad25015cc9b760a49bdb8e8fd26c87c2a359c | fix: workflows | [
{
"path": ".github/workflows/test.yaml",
"patch": "@@ -15,10 +15,13 @@ jobs:\n GENERATE_CUDA: ${{ steps.changes.outputs.GENERATE_CUDA }}\n GENERATE_ROCM: ${{ steps.changes.outputs.GENERATE_ROCM }}\n steps:\n+ - uses: actions/checkout@v4\n+ with:\n+ fetch-depth: 0\n ... | 2024-03-27T23:14:20 |
ggml-org/llama.cpp | f32ca51bfeb74b8d1b735c2744c0e4a6224f6b7c | e1f4921980444b30d3b4fa1c6e18cdecd85b0690 | server: add presets (config) when using multiple models (#17859)
* llama-server: recursive GGUF loading
Replace flat directory scan with recursive traversal using
std::filesystem::recursive_directory_iterator. Support for
nested vendor/model layouts (e.g. vendor/model/*.gguf).
Model name now reflects the relative pat... | [
{
"path": "common/CMakeLists.txt",
"patch": "@@ -73,6 +73,8 @@ add_library(${TARGET} STATIC\n ngram-cache.h\n peg-parser.cpp\n peg-parser.h\n+ preset.cpp\n+ preset.h\n regex-partial.cpp\n regex-partial.h\n sampling.cpp",
"additions": 2,
"deletions": 0,
"language": "... | 2025-12-10T21:18:21 |
denoland/deno | fce52fa2e9adcb236c4f01072e423601ec33b2f3 | 148e8a0b1c773b0fb219bf834d004be62120ae20 | fix(ext/node): support HTTP over Windows named pipes in node:http (#32414)
Add Windows Named Pipe support to Deno's Node.js HTTP compatibility
layer, enabling libraries like `dockerode`/`docker-modem` that use
`http.request({socketPath: "//./pipe/docker_engine"})` to work on
Windows.
Add `WindowsPipe` variant to `Net... | [
{
"path": "ext/http/http_next.rs",
"patch": "@@ -1007,6 +1007,10 @@ where\n NetworkStream::Tunnel(conn) => {\n serve_http(conn, connection_properties, lifetime, tx, options)\n }\n+ #[cfg(windows)]\n+ NetworkStream::WindowsPipe(conn) => {\n+ serve_http(conn, connection_properties, li... | 2026-03-04T10:11:29 |
ollama/ollama | af8a8a6b59080f2309e7761e995d7b83bfb8c941 | 461ad25015cc9b760a49bdb8e8fd26c87c2a359c | fix: trim quotes on OLLAMA_ORIGINS | [
{
"path": "server/routes.go",
"patch": "@@ -1013,16 +1013,14 @@ func allowedHostsMiddleware(addr net.Addr) gin.HandlerFunc {\n }\n \n func (s *Server) GenerateRoutes() http.Handler {\n-\tvar origins []string\n-\tif o := os.Getenv(\"OLLAMA_ORIGINS\"); o != \"\" {\n-\t\torigins = strings.Split(o, \",\")\n-\t}... | 2024-03-27T22:24:28 |
vuejs/vue | c2b1cfe9ccd08835f2d99f6ce60f67b4de55187f | 0496115c55b4539d1901ce409a457f8b9fbc20ce | fix(weex): donot rethrow the captured error on weex platform (#7024) | [
{
"path": "src/core/util/env.js",
"patch": "@@ -1,16 +1,19 @@\n /* @flow */\n+declare var WXEnvironment: any;\n \n // can we use __proto__?\n export const hasProto = '__proto__' in {}\n \n // Browser environment sniffing\n export const inBrowser = typeof window !== 'undefined'\n+export const inWeex = typeof... | 2017-11-13T16:48:41 |
ggml-org/llama.cpp | e1f4921980444b30d3b4fa1c6e18cdecd85b0690 | 4dff236a522bd0ed949331d6cb1ee2a1b3615c35 | Fix race conditions in threadpool when dealing with dynamic/frequent n_threads changes (#17748)
* tests: update barrier test to check for race condition in active threads
* cpu: combine n_graph and n_threads into a single atomic update
* tests: add multi-graph test for test_barrier | [
{
"path": "ggml/src/ggml-cpu/ggml-cpu.c",
"patch": "@@ -187,6 +187,9 @@ typedef void * thread_ret_t;\n \n typedef pthread_t ggml_thread_t;\n \n+#define GGML_THREADPOOL_N_THREADS_MASK (0xffffU)\n+#define GGML_THREADPOOL_N_THREADS_BITS (16)\n+\n #if defined(__APPLE__)\n #include <unistd.h>\n #include <mach/ma... | 2025-12-10T20:32:23 |
huggingface/transformers | 05c0e1d39082ee8b69064ed4ea9c239cd17405e9 | 01c51596ec56407eabc7f8d1f021a8aeb679a330 | rm slow tokenizers (#40936)
* fixes missed
* gemma test fix
* refactor
* rm legacy from llama
* added renaming
* add _model
* update legacy
* update legacy
* fix docstring
* always load blank, then set _tokenizer if we have it
* new toks
* update all berttokenizer based models
* apply feedback - delete ber... | [
{
"path": "MIGRATION_GUIDE_V5.md",
"patch": "@@ -74,6 +74,207 @@ While this is being implemented, expect varying levels of support across differe\n \n Linked PR: https://github.com/huggingface/transformers/pull/41580\n \n+\n+\n+\n+## Tokenization\n+\n+Just as we moved towards a single backend library for mo... | 2025-11-27T18:24:50 |
denoland/deno | 2334e53e96613878baaeac8ae4847e84e7c2ec36 | 86a97b288d337e30ece2739a571e1a0c26f1664e | ci: fix npm publish (#32437) | [
{
"path": ".github/workflows/npm_publish.yml",
"patch": "@@ -36,10 +36,8 @@ jobs:\n - name: Install Node\n uses: actions/setup-node@v6\n with:\n- node-version: '22.x'\n+ node-version: '24.x'\n registry-url: 'https://registry.npmjs.org'\n \n - name: Pub... | 2026-03-03T16:41:49 |
vuejs/vue | 9463ac8746f57d1dff6f958b7d79765aa0979fa8 | f0a66c5817f6580519c7f33704c8013398724039 | test: fix weex tests | [
{
"path": "test/weex/compiler/compile.spec.js",
"patch": "@@ -32,12 +32,12 @@ describe('compile basic', () => {\n \n it('should compile unary tag', () => {\n const inputCase = compile(`<div><input><text>abc</text></div>`)\n- expect(inputCase.render).toMatch(strToRegExp(`return _m(0)`))\n+ expect... | 2017-11-07T22:25:55 |
huggingface/transformers | 01c51596ec56407eabc7f8d1f021a8aeb679a330 | a099b27ddb2fe4a0f690ba6007ee38f109ccbf31 | update with more recent tts models (#42328)
* update with more recent tts models
* fix pipelin | [
{
"path": "docs/source/en/tasks/text-to-speech.md",
"patch": "@@ -19,18 +19,18 @@ rendered properly in your Markdown viewer.\n [[open-in-colab]]\n \n Text-to-speech (TTS) is the task of creating natural-sounding speech from text, where the speech can be generated in multiple\n-languages and for multiple spe... | 2025-11-27T17:59:43 |
ggml-org/llama.cpp | 6c2131773c5975c719863a21a8195a66dea4a4d0 | b6777218198888ed4b18d839e805f887562a9be4 | cli: new CLI experience (#17824)
* wip
* wip
* fix logging, add display info
* handle commands
* add args
* wip
* move old cli to llama-completion
* rm deprecation notice
* move server to a shared library
* move ci to llama-completion
* add loading animation
* add --show-timings arg
* add /read command, im... | [
{
"path": ".github/workflows/build.yml",
"patch": "@@ -243,7 +243,7 @@ jobs:\n echo \"Fetch llama2c model\"\n wget https://huggingface.co/karpathy/tinyllamas/resolve/main/stories260K/stories260K.bin\n ./bin/llama-convert-llama2c-to-ggml --copy-vocab-from-model ./tok512.bin --ll... | 2025-12-10T14:28:59 |
denoland/deno | 86a97b288d337e30ece2739a571e1a0c26f1664e | 3fe843d26be8060735b067954f9686a818362173 | test: dgram.addMembership EINVAL error regression test (#32256)
Actually fixed in #32381
Closes #31059
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> | [
{
"path": "tests/unit_node/dgram_test.ts",
"patch": "@@ -58,6 +58,20 @@ Deno.test(\"[node/dgram] udp unref\", {\n assertEquals(statusCode, 0);\n });\n \n+Deno.test(\"[node/dgram] addMembership works\", async () => {\n+ const { promise, resolve } = Promise.withResolvers<void>();\n+ const socket = createS... | 2026-03-03T16:06:31 |
vuejs/vue | 267ada04e8dd66f5c159dd6ba1b9f88fbbe78676 | 2acc12c9edb03329c4d9cddcca26e46c672a77bc | fix(types): expose VueConstructor (#7002) | [
{
"path": "types/index.d.ts",
"patch": "@@ -3,7 +3,8 @@ import { Vue } from \"./vue\";\n export default Vue;\n \n export {\n- CreateElement\n+ CreateElement,\n+ VueConstructor\n } from \"./vue\";\n \n export {",
"additions": 2,
"deletions": 1,
"language": "Unknown"
}
] | 2017-11-06T21:34:22 |
huggingface/transformers | a8457eb866c8fc6cf3b4db0eb51288b27c001f44 | e2e411c42b1fff2a5f988170def15ca47739fd15 | Fix the "test_offline" test (#42458)
* Fix the test offline test
* Refactor test_offline method with context patches
Refactor test_offline to use patch for offline mode.
* Apply suggestion from @Wauplin
---------
Co-authored-by: Lucain <lucainp@gmail.com> | [
{
"path": "tests/utils/test_modeling_utils.py",
"patch": "@@ -290,22 +290,20 @@ def forward(self, mask, inputs_embeds):\n return attention_mask\n \n class TestOffline(unittest.TestCase):\n- @patch(\"huggingface_hub.constants\")\n- def test_offline(self, mock_hf_hub_constants):\... | 2025-11-27T16:36:58 |
ggml-org/llama.cpp | 17f7f4baad8b3a716ee139da7bb56ae984e8c0fa | 9e79b0116ebb6ff4a1ef1b42a7f2f64182ec4f10 | CUDA: fix unpadded strides in MMA FA kernel (#17891) | [
{
"path": "ggml/src/ggml-cuda/fattn-mma-f16.cuh",
"patch": "@@ -955,43 +955,40 @@ static __device__ __forceinline__ void flash_attn_ext_f16_process_tile(\n (K_h2 + int64_t(kb0)*nbatch_fa*stride_K, tile_K, nbatch_K2, stride_K, k_VKQ_sup);\n }\n \n- for (; kb0 < kb0_stop-1; ++kb0) {\n- ... | 2025-12-10T11:39:56 |
ollama/ollama | b8c2be6142068b540406d2f9c84b1576b4394b0a | e0319bd78d834675a06fb2e9e2673efe29dfa722 | Use Rocky Linux Vault to get GCC 10.2 installed
This should hopefully only be a temporary workaround until Rocky 8
picks up GCC 10.4 which fixes the NVCC bug | [
{
"path": "scripts/rh_linux_deps.sh",
"patch": "@@ -20,7 +20,19 @@ if grep -i \"centos\" /etc/system-release >/dev/null; then\n fi\n dnf install -y devtoolset-10-gcc devtoolset-10-gcc-c++\n elif grep -i \"rocky\" /etc/system-release >/dev/null; then\n- dnf install -y git gcc-toolset-10-gcc gcc-to... | 2024-03-26T02:18:50 |
vuejs/vue | ad673e12575404ee68ed267d4186de803f3dbb08 | 5a9da95b8a865416f082952a48416ffc091e4078 | chore: fix flow | [
{
"path": "src/server/optimizing-compiler/optimizer.js",
"patch": "@@ -130,11 +130,11 @@ function hasCustomDirective (node: ASTNode): ?boolean {\n \n // <select v-model> cannot be optimized because it requires a runtime check\n // to determine proper selected option\n-function isSelectWithModel (node: ASTNo... | 2017-11-03T20:51:16 |
denoland/deno | 5d6b10607a0512f5bcac2158cefd110b259a9aac | 75c96382a90f7de074766d66112f17838aea7a6b | fix(ext/node): handle partial writes to process.stdout/stderr (#32433)
## Summary
- Fix silent data truncation when piping >64KB through `process.stdout`
in Node compat mode
- The `_write` handler in `createWritableStdioStream` discarded the
return value of `writer.writeSync()`, assuming all bytes were always
written... | [
{
"path": "ext/node/polyfills/_process/streams.mjs",
"patch": "@@ -47,11 +47,18 @@ export function createWritableStdioStream(writer, name, warmup = false) {\n // being created from a raw fd (new Socket({ fd: 1 })), process.stdout/stderr\n // should be switched to net.Socket for non-TTY cases and... | 2026-03-03T15:56:51 |
ggml-org/llama.cpp | 2e9eab80c26d4b7e64f27c48b5af683c35c28742 | 2fbe3b7bb7814621477d658a81f0d0cca08761cf | fix softmax for iGPU (#17838) | [
{
"path": "ggml/src/ggml-sycl/softmax.cpp",
"patch": "@@ -277,7 +277,7 @@ static void soft_max_f32_sycl(const float *x, const T *mask,\n const int id = get_current_device_id();\n const size_t smpbo = ggml_sycl_info().devices[id].smpbo;\n \n- if (nbytes_shared <= smpbo) {\n+ if (nbytes_sh... | 2025-12-10T08:59:57 |
huggingface/transformers | 9bdec656c992617b4798589faa3f7c8bf21bb30a | e31c5f603e3c9872372bdfe807b1e85e36631798 | Let's break Qwen-VL 🚨 (#42420)
* Lets braeaak!
* pop attributes so we don't have 2 different values for the same key. Reported by TRL in the past
* oh no, tying is messing up with slow tests....
* fix copies
* migration guide | [
{
"path": "MIGRATION_GUIDE_V5.md",
"patch": "@@ -128,7 +128,7 @@ model_4bit = AutoModelForCausalLM.from_pretrained(\n - Methods to init a nested config such as `from_xxx_config` are deleted. Configs can be init from the `__init__` method in the same way. See [#41314](https://github.com/huggingface/transform... | 2025-11-27T15:25:16 |
vuejs/vue | 5a9da95b8a865416f082952a48416ffc091e4078 | 4fe1a95d2953ecf765e27677fa70ebadb176d4c3 | fix(slots): properly handle nested named slot passing
fix #6996 | [
{
"path": "src/compiler/parser/index.js",
"patch": "@@ -467,7 +467,7 @@ function processSlot (el) {\n el.slotTarget = slotTarget === '\"\"' ? '\"default\"' : slotTarget\n // preserve slot as an attribute for native shadow DOM compat\n // only for non-scoped slots.\n- if (!el.slotScope... | 2017-11-03T20:45:54 |
denoland/deno | d45142b6d82b26945a6b43e0e57355a3a8631888 | 83e374686079a2f067115e60ad91ec38f4f6118b | fix(ext/node): make UDP socket ops synchronous, implement missing dgram methods (#32381)
## Summary
- Creates a new `NodeUdpSocketResource` in `ext/node` that stores
`tokio::net::UdpSocket` directly (no `AsyncRefCell`), making all socket
config operations synchronous
- Replaces `udp_wrap.ts`'s dependency on `ext:deno... | [
{
"path": "ext/node/lib.rs",
"patch": "@@ -337,6 +337,20 @@ deno_core::extension!(deno_node,\n ops::inspector::op_inspector_disconnect,\n ops::inspector::op_inspector_emit_protocol_event,\n ops::inspector::op_inspector_enabled,\n+ ops::udp::op_node_udp_bind,\n+ ops::udp::op_node_udp_join_m... | 2026-03-03T14:55:23 |
ggml-org/llama.cpp | 086a63e3a5d2dbbb7183a74db453459e544eb55a | b63509262ac88db248fa70a1364c8a76e339e93f | metal: SSM kernel improvements (#17876)
* feat: Add a batched version of ssm_conv
This was done using Claude Code. It found a number of optimizations around
how the threads were organized, resulting in a huge performance boost!
Branch: Mamba2SSD
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* feat: Optimized SSM... | [
{
"path": "ggml/src/ggml-metal/ggml-metal-device.cpp",
"patch": "@@ -411,6 +411,38 @@ ggml_metal_pipeline_with_params ggml_metal_library_get_pipeline_ssm_conv(ggml_me\n return res;\n }\n \n+ggml_metal_pipeline_with_params ggml_metal_library_get_pipeline_ssm_conv_batched(ggml_metal_library_t lib, const g... | 2025-12-09T19:30:02 |
huggingface/transformers | e31c5f603e3c9872372bdfe807b1e85e36631798 | 52c5c6582bdef8bd0f0a9238c9d6703137a10583 | FIX: Minimal fix for loading PEFT weights (#42387)
* FIX Minimal fix for loading PEFT weights
After the weight conversion PR #41580, some adjustments were still
required for loading PEFT weights. This PR presents a minimal fix to
make it work again.
Besides renaming keys, this PR does not address possible conversion... | [
{
"path": "src/transformers/integrations/peft.py",
"patch": "@@ -15,9 +15,9 @@\n import inspect\n import json\n import os\n-import re\n from typing import Any, Literal\n \n+from ..core_model_loading import WeightRenaming, build_glob_alternation, repl\n from ..utils import (\n CONFIG_NAME,\n cached_f... | 2025-11-27T15:13:44 |
vuejs/vue | 4fe1a95d2953ecf765e27677fa70ebadb176d4c3 | e1657fd7ce49bff3c3ecad3c56ae527347505c34 | fix: properly mark slot rendered flag in production mode
fix #6997 | [
{
"path": "src/core/instance/render-helpers/render-slot.js",
"patch": "@@ -27,12 +27,14 @@ export function renderSlot (\n } else {\n const slotNodes = this.$slots[name]\n // warn duplicate slot usage\n- if (slotNodes && process.env.NODE_ENV !== 'production') {\n- slotNodes._rendered && war... | 2017-11-03T19:40:40 |
denoland/deno | c15deead44dcc7208a045ba0503e2bf8e9cf05cf | 3f459413b2d18458268024d1ce0dad715240c2ce | fix(npm): NPM_CONFIG_REGISTRY correctly overrides .npmrc (#32394)
Co-authored-by: David Sherret <dsherret@gmail.com> | [
{
"path": "Cargo.lock",
"patch": "@@ -2779,6 +2779,7 @@ dependencies = [\n \"serde\",\n \"serde_json\",\n \"sha2\",\n+ \"sys_traits\",\n \"thiserror 2.0.12\",\n \"tokio\",\n \"url\",",
"additions": 1,
"deletions": 0,
"language": "Unknown"
},
{
"path": "cli/lsp/npm.rs",
"patch":... | 2026-03-03T12:13:11 |
ollama/ollama | f38b705dc7b18817020c0ae19cbb58d915e67324 | 560be5e0b60741600140fa0918ee125f0b7f9c75 | Fix ROCm link in `development.md` | [
{
"path": "docs/development.md",
"patch": "@@ -69,7 +69,7 @@ go build .\n \n _Your operating system distribution may already have packages for AMD ROCm and CLBlast. Distro packages are often preferable, but instructions are distro-specific. Please consult distro-specific docs for dependencies if available!_... | 2024-03-25T20:32:44 |
ggml-org/llama.cpp | 02e409a5becd2142c97a843bb63876fb8da7c6f0 | 6b82eb7883d1babf5f6bd4bca70997f229691fed | ggml : Provide macos-specific backtrace printing to avoid terminal death (#17869)
* fix: Provide macos-specific backtrace printing to avoid terminal death
Branch: MacOSSafeBacktrace
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
* fix: Add GGML_BACKTRACE_LLDB env var to enable using lldb for backtrace
Branch: Mac... | [
{
"path": "ggml/src/ggml.c",
"patch": "@@ -124,6 +124,13 @@ static void ggml_print_backtrace_symbols(void) {\n int nptrs = backtrace(trace, sizeof(trace)/sizeof(trace[0]));\n backtrace_symbols_fd(trace, nptrs, STDERR_FILENO);\n }\n+#elif defined(__APPLE__)\n+#include <execinfo.h>\n+static void ggml_... | 2025-12-09T16:29:07 |
huggingface/transformers | 52c5c6582bdef8bd0f0a9238c9d6703137a10583 | f624084ef479c9d449577314e9bde320f708bdeb | Correctly return finish reason length when finished (#42157)
* Correctly return finish reason length when finished
* Typos + fixup
* Fix a few tests
* Update src/transformers/cli/chat.py
Co-authored-by: Lucain <lucainp@gmail.com>
---------
Co-authored-by: Lucain <lucainp@gmail.com> | [
{
"path": "src/transformers/cli/chat.py",
"patch": "@@ -19,7 +19,7 @@\n import string\n import time\n from collections.abc import AsyncIterator\n-from typing import Annotated\n+from typing import Annotated, Any, Optional\n \n import click\n import typer\n@@ -103,12 +103,14 @@ def __init__(self, model_id: st... | 2025-11-27T14:55:55 |
denoland/deno | 74cc882fe924caeaea9abc7604dca76e4e83b980 | 0cfe40a82c91b3dfb8a3b4a818dbf190156bbbf7 | fix(ext/node): gracefully exit on invalid NODE_CHANNEL_FD instead of panicking (#32421)
## Summary
- When process managers like PM2 set `NODE_CHANNEL_FD` to a file
descriptor that isn't a valid IPC pipe, Deno would panic during
bootstrap with "Bootstrap exception"
- Now it prints a clear error message and exits with c... | [
{
"path": "ext/node/ops/ipc.rs",
"patch": "@@ -186,18 +186,28 @@ mod impl_ {\n log::debug!(\"op_node_child_ipc_pipe: {:?}, {:?}\", fd, serialization);\n let ref_tracker = IpcRefTracker::new(state.external_ops_tracker.clone());\n match serialization {\n- ChildIpcSerialization::Json => Ok(Som... | 2026-03-03T10:46:09 |
vuejs/vue | e1657fd7ce49bff3c3ecad3c56ae527347505c34 | e80104eb63ae44e159061a3753aa0356ac8b82eb | fix(ssr): properly render <select v-model> initial state
fix #6986 | [
{
"path": "src/platforms/web/server/directives/index.js",
"patch": "@@ -1,5 +1,7 @@\n import show from './show'\n+import model from './model'\n \n export default {\n- show\n+ show,\n+ model\n }",
"additions": 3,
"deletions": 1,
"language": "JavaScript"
},
{
"path": "src/platforms/web/... | 2017-11-03T19:30:47 |
ollama/ollama | 7b6cbc10ec9765b0013c163e868c5307d9d27009 | acfa2b94220882ffe8246a56e35a6cd1db38e6a2 | Integration tests conditionally pull
If images aren't present, pull them.
Also fixes the expected responses | [
{
"path": "integration/basic_test.go",
"patch": "@@ -12,7 +12,7 @@ import (\n )\n \n func TestOrcaMiniBlueSky(t *testing.T) {\n-\tctx, cancel := context.WithTimeout(context.Background(), 1*time.Minute)\n+\tctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)\n \tdefer cancel()\n \t// Set ... | 2024-03-24T23:22:38 |
ggml-org/llama.cpp | 6b82eb7883d1babf5f6bd4bca70997f229691fed | 86a3f0fad8b153ac9396e1ac18e790e4179c53f2 | metal : print node names for debugging (#17882) | [
{
"path": "ggml/src/ggml-metal/ggml-metal-ops.cpp",
"patch": "@@ -221,7 +221,7 @@ static int ggml_metal_op_encode_impl(ggml_metal_op_t ctx, int idx) {\n }\n \n if (ctx->debug_graph > 0) {\n- GGML_LOG_DEBUG(\"%s: node[%5d] - %-12s %s\\n\", __func__, idx, ggml_op_name(node->op), is_... | 2025-12-09T13:25:49 |
denoland/deno | 25dbe4c8b95a1c7bc6cee5880d63f6e73c117a78 | 28ec7c0ecdaf2648a8796f8766adac9ab09231e6 | fix(ext/node): support SEC1 EC private key import/export in `node:crypto` (#32401)
## Summary
- Fix SEC1 EC private key import failing with "malformed or missing
named curve in ec parameters" when the optional `parameters` field is
absent
- Fix SEC1 EC private key export to include the curve OID in the
`parameters` f... | [
{
"path": "ext/node_crypto/keys.rs",
"patch": "@@ -719,6 +719,50 @@ pub enum AsymmetricPublicKeyError {\n UnsupportedPrivateKeyOid,\n }\n \n+/// Parse an EC private key from SEC1 DER bytes using the named curve OID.\n+/// The curve OID is required — inferring from key length is unreliable\n+/// (e.g. P-25... | 2026-03-03T10:01:35 |
huggingface/transformers | 1bca9bac88b98b8809cc2d9c145c93ef1ec8a7da | 663b4cf121d2ab284bb24dd8174e53a79c404d15 | fix: Restore explicit .keys() calls for TensorDict compatibility (#42373)
* fix: Restore explicit .keys() calls for TensorDict compatibility
Fixes issue where TensorDict objects cause RuntimeError: generator raised StopIteration
when used with data collators and tokenization utilities.
Problem:
- TensorDict.__iter__... | [
{
"path": "src/transformers/feature_extraction_sequence_utils.py",
"patch": "@@ -123,8 +123,9 @@ def pad(\n # If we have a list of dicts, let's convert it in a dict of lists\n # We do this to allow using this method as a collate_fn function in PyTorch Dataloader\n if isinstance(proce... | 2025-11-27T14:31:55 |
vuejs/vue | 79c0d7bcfbcd1ac492e7ceb77f5024d09efdc6b3 | 53b77f805475f217eb221a475409e9a27f590549 | fix(ssr): properly render textarea value
partial fix for #6986 | [
{
"path": "src/platforms/web/server/modules/dom-props.js",
"patch": "@@ -27,7 +27,10 @@ export default function renderDOMProps (node: VNodeWithData): string {\n setText(node, props[key], true)\n } else if (key === 'textContent') {\n setText(node, props[key], false)\n+ } else if (key === '... | 2017-11-03T03:04:31 |
ollama/ollama | 6ce37e4d969b65e9af8e7b3bd3989412daf5bee5 | 703684a82a150ba59b300ca9c1ec678bd9b4ab27 | llm,readline: use errors.Is instead of simple == check (#3161)
This fixes some brittle, simple equality checks to use errors.Is. Since
go1.13, errors.Is is the idiomatic way to check for errors.
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com> | [
{
"path": "llm/gguf.go",
"patch": "@@ -3,6 +3,7 @@ package llm\n import (\n \t\"bytes\"\n \t\"encoding/binary\"\n+\t\"errors\"\n \t\"fmt\"\n \t\"io\"\n \t\"log/slog\"\n@@ -540,7 +541,7 @@ func (llm *GGUFModel) Encode(f *os.File) error {\n \t\t\tb, err := io.ReadFull(dataFile, data)\n \t\t\tremaining -= uint... | 2024-03-15T14:14:12 |
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.