1251 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Peilin Li	07e274467a	[build]: flatten ktransformers package shim (#1955)	2026-04-25 22:08:52 +08:00
Peilin Li	bfbd0e9352	[chore]: archive kt-sft package (#1954)	2026-04-25 21:49:21 +08:00
Peilin Li	85f1ab530b	[ci]: use hosted runner for sglang-kt release ... Use GitHub-hosted runners for the pure Python sglang-kt release workflow so the PyPI release is not blocked by unavailable self-hosted runners.	2026-04-25 21:05:18 +08:00
Peilin Li	bc7afff13b	[chore]: sync sglang-kt packaging fix ... Update third_party/sglang to the merged sglang-kt 0.6.1 dependency metadata fix so the release workflow builds the corrected inference package.	2026-04-25 21:02:25 +08:00
Peilin Li	eeaeb7bfd7	[build]: align kt-kernel torch support with v0.6.1 release (#1948)	2026-04-24 23:45:15 +08:00
Peilin Li	85308615b9	[build] prepare v0.6.1 SFT wheel packaging on main (#1945) ... * [build]: prepare 0.6.1 SFT wheel packaging on main * [build]: finalize py311+ wheel packaging defaults	2026-04-24 12:08:38 +08:00
mrhaoxx	9544a8960d	feat(sft): AMX MoE SFT backend with LoRA support (#1936) ... * feat(sft): AMX MoE SFT backend with LoRA support Complete SFT (Supervised Fine-Tuning) backend for MoE models using AMX SIMD: Core C++ implementation: - sft_moe.hpp: Forward/backward with LoRA fused operations (~5500 lines) - moe-sft-tp.hpp: Tensor-parallel wrapper for multi-NUMA - amx/moe-sft-tp.hpp: AMX-specific TP implementation - avx_kernels.hpp: AVX512 SIMD kernels for LoRA GEMM - amx_kernels.hpp: AMX tile kernels for Panel5 rank-outer optimization - worker_pool: RDTSC profiling, Chrome trace output, SFT timer infrastructure - ext_bindings.cpp: SFT MOE pybind bindings (BF16/INT8/INT4 + SkipLoRA variants) Python sft/ submodule (kt_kernel.sft): - base.py: BaseSFTMoEWrapper with buffer management (template method pattern) - amx.py: AMXSFTMoEWrapper (weight loading, C++ task construction) - autograd.py: KTMoEFunction (torch.autograd.Function for distributed training) - layer.py: KTMoELayerWrapper (nn.Module replacing HF MoE layers) - arch.py: MOEArchConfig (Qwen3/DeepSeek/Mixtral architecture detection) - weights.py: Expert weight extraction and checkpoint loading - lora.py: PEFT LoRA adaptation (view buffers, grad buffers, save/load adapter) - wrapper.py: wrap_moe_layers_with_kt_wrapper, load_kt_model, build_kt_device_map - config.py: KTConfig dataclass (DeepSpeed-style opaque config passthrough) - dist_utils.py: Distributed gather/scatter, checkpoint-phase detection Design decisions: - Rank-0-only expert pattern: only rank 0 holds C++ wrapper and expert weights - DeepSpeed-style integration: accelerate keeps only KTransformersPlugin (framework interaction fields), all logic in kt_kernel.sft - Inference isolation: importing kt_kernel does not load sft/ submodule - Old field name compatibility: _get_kt_config() converts kt_xxx→xxx automatically Verified: Qwen3-235B-A22B 4GPU AMXBF16 training, loss converges normally. * refactor(sft): unify KTConfig field names with kt_ prefix, add share_cache_pool, remove dead code - KTConfig fields all use kt_ prefix matching dict keys — eliminates _OLD_TO_NEW mapping and prefix-stripping in wrapper.py - Add kt_share_cache_pool field, auto-enabled when gradient_checkpointing is on (via training_args.py), flows through to C++ cache allocation - Remove dead checkpoint detection code: in_ckpt_recompute, in_ckpt_first_forward vars (assigned but never read), fallback _is_in_checkpoint_first_forward() function, unused inspect import - Remove redundant env var fallbacks in wrapper.py for share_backward_bb and share_cache_pool (KTConfig.__post_init__ already handles env vars) - Simplify layer.py checkpoint logic to single _checkpoint_hook_mode() check Verified: Qwen3-235B 3-step training on sap4, loss matches baseline (1.2886 / 1.9824 / 1.377 vs 1.2886 / 1.9766 / 1.3809) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * refactor(sft): share_backward_bb default True, share_cache_pool auto-derived - kt_share_backward_bb defaults to True (always saves memory) - kt_share_cache_pool no longer reads from env var; defaults False, auto-set to True by trainer_config_process when gradient checkpointing is enabled Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: add missing gpu_experts_mask=None to KTMoEWrapper call in SFT wrapper KTMoEWrapper.__new__() requires gpu_experts_mask as a positional argument, but the SFT wrapper omitted it, causing MoE layer wrapping to fail silently and FSDP2 to attempt broadcasting all expert weights (OOM/NCCL crash). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat(sft): support transformers v5 fused expert format Fused experts (e.g. Qwen3MoeExperts) store weights as 3D Parameters (gate_up_proj [E,2I,H], down_proj [E,H,I]) instead of per-expert nn.Linear modules. PEFT cannot attach LoRA to these, so we create KT-managed LoRA buffers with kaiming init, nn.Parameter wrappers for the optimizer, and pre-assigned .grad for C++ backward. - arch.py: detect_fused_experts() detection - weights.py: fused format extraction and weight clearing - wrapper.py: detect fused at wrap time, store _fused_experts/_lora_rank - lora.py: _create_fused_expert_lora_buffers, save/load fused LoRA, get_kt_lora_params collects fused params, deduplicate wrapper finding - layer.py: handle v5 TopKRouter tuple output, remove dead code - autograd.py: sync_forward_sft/submit_forward_sft API rename Verified: v5 loss/expert-LoRA values match v4 baseline, v4 backward compat. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat(sft): add Qwen3.5 MoE support + fused checkpoint loading - arch.py: add Qwen3_5Moe arch match, read config from text_config, _get_layers_prefix returns model.language_model.layers for Qwen3.5, _get_model_container_and_layers searches language_model attr - weights.py: load_experts_from_checkpoint_files detects fused format (gate_up_proj in weight_map) and splits into gate/up/down - wrapper.py: hidden_size fallback to text_config Verified: Qwen3.5-35B-A3B (256 experts, fused format) E2E pass. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * [fix](sft): align Python API with C++ backend after v5 refactor - wrapper.py: pass gpu_experts_mask=None to KTMoEWrapper (required by C++ signature) - layer.py: rename submit_forward_sft/sync_forward_sft to submit_forward/sync_forward - autograd.py: rename sync_forward_sft to sync_forward The sft-v5 refactor (commits 58d7eab, dd1da65) renamed Python-side method calls but the C++ backend (AMXSFTMoEWrapper) still exposes the original method names. This caused AttributeError on Qwen3.5-35B and other models. * align sft branch with main: revert worker_pool, strip sft_timer, fix inference defaults - Revert worker_pool.cpp/.h to main (remove RDTSC timer, Chrome Trace, sft_timer namespace, ITT API, extended do_work_stealing_job API) - Strip all sft_timer instrumentation from sft-only files (sft_moe.hpp, moe-sft-tp.hpp, avx_kernels.hpp) - Restore pin_memory=True in KExpertsCPUBuffer (inference path) - Restore fused tensor transpose logic in convert_cpu_weights.py (main layout) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * revert CMakeLists.txt to main: remove debug flags and cpptrace dep Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * clean up dev artifacts: remove SFT design docs, debug examples, bench scripts Remove files not needed in the merge: - docs/SFT+KTWrapper/ (6 Chinese design docs) - docs/sft_moe_amx/ (21 dev/debug docs) - 12 debug/test example scripts - 6 SFT-specific bench scripts and report Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * remove dev version stamps from ext_bindings, sft_moe, moe-sft-tp Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: JimmyPeilinLi <lipeilin@mail.nwpu.edu.cn>	2026-04-22 11:27:01 +08:00
ErvinXie	22e9915ec9	docs: add GOSIM 2026 announcement and update roadmap link to Q2 (#1937) ... Co-authored-by: xwy <xwy@CHININLAIdeMacBook-Air-2.local>	2026-04-21 15:51:08 +08:00
Jianwei Dong	e327db58be	Update README.md (#1935)	2026-04-18 21:30:13 +08:00
Jianwei Dong	a9f28d495b	Update README.md (#1934)	2026-04-18 21:10:25 +08:00
Benjamin F	06ee9f62f3	[doc]: add prerequisite note for GLM-5.1 tutorial (#1932)	2026-04-14 15:07:08 +08:00
callmegaga	a9411f1d72	Supports vnni-256 for GPTQ INT4 (#1926) ... * [feat](kt-kernel): support avx-vnni-256 for gptq int4	2026-04-13 17:59:59 +08:00
Andy18650	f42e94a527	[fix](cli): handle edge cases with empty NUMA nodes (#1929) ... Co-authored-by: Andy18650 <114562805@qq.com>	2026-04-13 16:45:41 +08:00
ErvinXie	279c920a69	Revert "kt-kernel: enable CPUInfer stream bridge for ROCm (#1918)" (#1925) ... This reverts commit 1dd0a78899.	2026-04-09 18:43:03 +08:00
guanjiawei	1dd0a78899	kt-kernel: enable CPUInfer stream bridge for ROCm (#1918)	2026-04-09 12:20:04 +08:00
acture	9b2d3b687b	fix: remove broken symlink in archive/ktransformers/ (#1906) ... The symlink `archive/ktransformers/ktransformers` points to `/home/djw/py311_717/ktransformers/ktransformers`, an absolute path on a developer's local machine. It was introduced in #1581 during the repository restructuring and is broken on every other machine. Tools that recursively copy the repo tree (e.g. shutil.copytree) fail with FileNotFoundError on this dangling link.	2026-04-09 11:42:19 +08:00
Oql	ad19a3e653	Chore/kt layerwise prefill main (#1920) ... * [chore]: sync sglang submodule with KT layerwise prefill log rename * [chore]: use sglang main with KT layerwise prefill logs	2026-04-09 11:28:37 +08:00
Jianwei Dong	891c5c0a13	Support glm5.1 (#1916)	2026-04-07 11:29:32 +08:00
Jim James	8a427c9321	[feat]: add AVX512F+BW fallback for FP8 and BF16 under AMX backend (#1908)	2026-04-03 12:46:22 +08:00
Jianwei Dong	db9326302b	chore: bump version to 0.5.3 (#1909)	2026-04-01 18:58:48 +08:00
Oql	9e6484a538	[fix]: fix --numa-nodes handling (#1904) ... * [fix]: fix --numa-nodes handling	2026-03-31 17:50:22 +08:00
Oql	cdc867c864	[chore]: sync sglang submodule (#1903)	2026-03-31 11:21:09 +08:00
Doctor Shotgun	24cd4fc055	feat(kt-kernel): Add utility script to merge loose layer weights to safetensors (#1886) ... * Add utility script to merge loose layer weights to safetensors * Send warnings and errors to stderr * Fix expert index parsing for MOE_INT4 and MOE_INT8	2026-03-31 10:41:07 +08:00
alin899992	9c18b60556	feat: CPU weight conversion for GLM-5 and MiniMax-M2.5 (#1853) ... * Support for GLM-5 and Minimax-M2.5 Add CPU weight conversion support for GLM-5 and Minimax-M2.5 * fix: remove overly restrictive MiniMax condition and deduplicate code - Remove `args.input_type == "fp8"` from MiniMaxConverter selection so bf16/fp16 MiniMax models no longer fall through to OnlineQuantConverter (which doesn't handle w1/w2/w3 naming and would fail). - Remove OnlineQuantConverter._find_expert_layers() which is identical to the inherited ConverterBase._find_expert_layers(). - Remove redundant expert_key_filter assignment (same as base default). --------- Co-authored-by: ErvinXie <ervinxie@foxmail.com>	2026-03-31 10:39:48 +08:00
ErvinXie	3903c9afcc	(kt-kernel): add numa_nodes parameter for explicit NUMA node mapping (#1891) ... Add numa_nodes parameter to BaseMoEWrapper and all subclasses, allowing users to explicitly specify which NUMA node IDs to use for subpool mapping instead of always defaulting to sequential [0, 1, ..., N-1]. This enables running multiple KTransformers instances on different NUMA nodes of the same machine, e.g. --kt-threadpool-count 1 --kt-numa-nodes 1 to bind to NUMA node 1. Previously this required external numactl workarounds since subpool_numa_map was hardcoded to start from 0.	2026-03-31 10:27:50 +08:00
Doctor Shotgun	bdf4bb76c5	Fix worker pool idle CPU usage (#1902) ... * Replace sleep polling with condition variable blocking	2026-03-30 20:29:17 +08:00
benapple [msft]	7a02daa694	Fix TaskQueue worker thread 100% CPU spin when idle (#1899) ... * initial fix for issue 1858 * [fix]: add done flag check to sync() wait predicate to prevent deadlock during destruction Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Ben Appleby <Ben.Appleby@microsoft.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-03-27 23:20:42 +08:00
mrhaoxx	7a9daf0cd4	[feat](kt-kernel): support avx2 only inference for bf16 fp8 and gptq int4 (#1892) ... * feat: support avx2 bf16 fp8 inference * feat: support avx2 gptq int4 inference * fix: numeric issues in fp8 dequant * Tutorial avx2 (#1900) * fix: prevent injecting -DLLAMA_AVX512=ON on AVX2-only machines * docs: add AVX2 tutorial for running KTransformers on AVX2-only CPUs * Tutorial avx2 (#1901) * fix: prevent injecting -DLLAMA_AVX512=ON on AVX2-only machines * docs: add AVX2 tutorial for running KTransformers on AVX2-only CPUs * docs: update README.md --------- Co-authored-by: Benjamin F <159887351+yyj6666667@users.noreply.github.com>	2026-03-27 14:45:02 +08:00
YIFANCHENGDU	8561a71dd1	[fix] improve Sglang kt-kernel detect time duration (#1887) ... * Increase timeout for Check if --kt-gpu-prefill-token-threshold is in the help output to 90 seconds. In cloud environments,CUDA initialization and Python module loading can easily exceed 30 seconds. * Update kt-kernel/python/cli/utils/sglang_checker.py add comment about the change Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> --------- Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>	2026-03-18 23:07:40 +08:00
github-actions[bot]	7a4b9b0e87	[build]: sync sglang submodule to f6adb4f473ba9a767cd60237ef8325cfcd97eba9 (#1876) ... Co-authored-by: ovowei <80044717+ovowei@users.noreply.github.com>	2026-03-04 17:18:08 +08:00
Jianwei Dong	15c624dcae	Fix/sglang kt detection (#1875) ... * [feat]: simplify sglang installation with submodule, auto-sync CI, and version alignment - Add kvcache-ai/sglang as git submodule at third_party/sglang (branch = main) - Add top-level install.sh for one-click source installation (sglang + kt-kernel) - Add sglang-kt as hard dependency in kt-kernel/pyproject.toml - Add CI workflow to auto-sync sglang submodule daily and create PR - Add CI workflow to build and publish sglang-kt to PyPI - Integrate sglang-kt build into release-pypi.yml (version.py bump publishes both packages) - Align sglang-kt version with ktransformers via SGLANG_KT_VERSION env var injection - Update Dockerfile to use submodule and inject aligned version - Update all 13 doc files, CLI hints, and i18n strings to reference new install methods Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * [build]: bump version to 0.5.2 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * [build]: rename PyPI package from kt-kernel to ktransformers Users can now `pip install ktransformers` to get everything (sglang-kt is auto-installed as a dependency). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Revert "[build]: rename PyPI package from kt-kernel to ktransformers" This reverts commit e0cbbf6364. * [build]: add ktransformers meta-package for PyPI `pip install ktransformers` now works as a single install command. It pulls kt-kernel (which in turn pulls sglang-kt). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * [fix]: show sglang-kt package version in kt version command - Prioritize sglang-kt package version (aligned with ktransformers) over sglang internal __version__ - Update display name from "sglang" to "sglang-kt" Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * [fix]: improve sglang-kt detection in kt doctor and kt version Recognize sglang-kt package name as proof of kvcache-ai fork installation. Previously both commands fell through to "PyPI (not recommended)" for non-editable local source installs. Now version.py reuses the centralized check_sglang_installation() logic. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * [build]: bump version to 0.5.2.post1 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-04 16:54:48 +08:00
Chen Hongtao	9e69fccb02	[feat]: add mistral moe loader compatibility (#1873) ... Co-authored-by: chenht2022 <chenht2022@users.noreply.github.com>	2026-02-28 17:50:23 +08:00
Jianwei Dong	19887e4363	update docker build (#1872)	2026-02-28 10:34:35 +08:00
VYSE V.E.O	20262b2743	Fix Qwen3.5 FP8 load for VL detection (#1857) ... * Fix Qwen3.5 FP8 load for VL detection 1, for VL models(Qwen3.5), modify base_key: model.layers.{N} -> model.language_model.layers.{N} 2, clean DUPLICATED class BF16SafeTensorLoader(SafeTensorLoader) , only the first overrided one. * Indent type Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> --------- Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>	2026-02-26 15:47:22 +08:00
Rin	786987a95f	Handle unquoted paths and special characters in model scanner (#1840) ... * Handle unquoted paths and special characters in model scanner * Fix ValueError: capture_output cannot be used with stderr `capture_output=True` internally sets `stderr=PIPE`, which conflicts with `stderr=subprocess.DEVNULL`. Replace `capture_output=True` with explicit `stdout=subprocess.PIPE` to keep stderr suppressed correctly. Also remove redundant `shell=False` (already the default). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: ErvinXie <ervinxie@foxmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-26 15:44:45 +08:00
Jianwei Dong	16a8b98f3e	support qwen3.5 (#1846)	2026-02-16 15:48:14 +08:00
Jiaqi Liao	411b69bec0	Add GLM-5 Day0 Support update to README (#1851)	2026-02-15 11:37:12 +08:00
Jiaqi Liao	7d9943365a	Add MiniMax-M2.5 Day0 support update (#1850) ... Added update for MiniMax-M2.5 Day0 support in the README.	2026-02-13 22:49:10 +08:00
Jiaqi Liao	a3d5d53605	Update MiniMax-M2.5.md (#1849)	2026-02-13 22:35:36 +08:00
Jianwei Dong	f0e4fc612b	support minimax-m2.5 (#1848)	2026-02-13 19:15:44 +08:00
Oql	1c72b3f5bd	fix glm5 docs (#1845)	2026-02-12 02:33:37 +08:00
Oql	7f7aeaeff6	support glm 5 (#1844)	2026-02-12 02:03:32 +08:00
Jiaqi Liao	061fb56382	Update Kimi-K2.5.md (#1838)	2026-02-07 16:38:39 +08:00
Chen Hongtao	d342fb1df6	[docs]: add maintainers list (#1837) ... * [docs]: add maintainers list * Apply suggestion from @Copilot Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --------- Co-authored-by: ErvinXie <ervinxie@qq.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>	2026-02-06 18:28:06 +08:00
Oql	56cbd69ac4	kt-cli enhancement (#1834) ... * [feat]: redesign kt run interactive configuration with i18n support - Redesign kt run with 8-step interactive flow (model selection, inference method, NUMA/CPU, GPU experts, KV cache, GPU/TP selection, parsers, host/port) - Add configuration save/load system (~/.ktransformers/run_configs.yaml) - Add i18n support for kt chat (en/zh translations) - Add universal input validators with auto-retry and Chinese comma support - Add port availability checker with auto-suggestion - Add parser configuration (--tool-call-parser, --reasoning-parser) - Remove tuna command and clean up redundant files - Fix: variable reference bug in run.py, filter to show only MoE models * [feat]: unify model selection UI and enable shared experts fusion by default - Unify kt run model selection table with kt model list display * Add Total size, MoE Size, Repo, and SHA256 status columns * Use consistent formatting and styling * Improve user decision-making with more information - Enable --disable-shared-experts-fusion by default * Change default value from False to True * Users can still override with --enable-shared-experts-fusion * [feat]: improve kt chat with performance metrics and better CJK support - Add performance metrics display after each response * Total time, TTFT (Time To First Token), TPOT (Time Per Output Token) * Accurate input/output token counts using model tokenizer * Fallback to estimation if tokenizer unavailable * Metrics shown in dim style (not prominent) - Fix Chinese character input issues * Replace Prompt.ask() with console.input() for better CJK support * Fixes backspace deletion showing half-characters - Suppress NumPy subnormal warnings * Filter "The value of the smallest subnormal" warnings * Cleaner CLI output on certain hardware environments * [fix]: correct TTFT measurement in kt chat - Move start_time initialization before API call - Previously start_time was set when receiving first chunk, causing TTFT ≈ 0ms - Now correctly measures time from request sent to first token received * [docs]: 添加 Clawdbot 集成指南 - KTransformers 企业级 AI 助手部署方案 * [docs]: 强调推荐使用 Kimi K2.5 作为核心模型，突出企业级推理能力 * [docs]: 添加 Clawdbot 飞书接入教程链接 * [feat]: improve CLI table display, model verification, and chat experience - Add sequence number (#) column to all model tables by default - Filter kt edit to show only MoE GPU models (exclude AMX) - Extend kt model verify to check *.json and *.py files in addition to weights - Fix re-verification bug where repaired files caused false failures - Suppress tokenizer debug output in kt chat token counting * [fix]: fix cpu cores. --------- Co-authored-by: skqliao <skqliao@gmail.com>	2026-02-04 16:44:54 +08:00
Oql	4f64665758	[docs]: add Qwen3 Coder Next Tutorial (#1833)	2026-02-04 16:27:10 +08:00
Oql	c28cfcb26e	[fix]: fix k2-moe.hpp load weight (#1830)	2026-02-03 11:28:49 +08:00
Jiaqi Liao	794c04fae4	Revert "[doc]: update kimi_k2.5 doc (#1823)" (#1825) ... This reverts commit 2e6506535b.	2026-01-30 16:10:01 +08:00
Oql	ccbb5b1cf8	[docs]: add clawd bot docs ... add clawd bot docs	2026-01-30 15:51:30 +08:00
Jiaqi Liao	2e6506535b	[doc]: update kimi_k2.5 doc (#1823)	2026-01-30 15:43:18 +08:00