intel neural-speed issues

intel / neural-speed

An innovative library for efficient LLM inference via low-bit quantization

https://github.com/intel/neural-speed

Apache License 2.0

349 stars 38 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Once upon a time, a little NE_ASSERT: /root/w0/workspace/neuralspeed-wheel-build/nlp_repo/neural_speed/core/ne_layers.c:2651: ne_nelements(a) == ne0 * ne1 * ne2

#326 zwx109473 opened 3 months ago
2
XeTLA Support Compiler 2025.0

#325 DDEle closed 3 months ago
1
XeTLA Use ESIMD 2D Load APIs

#324 DDEle opened 3 months ago
0
Revert max_load_vec_elems

#323 DDEle closed 3 months ago
2
Xetla Fix GRF Mode Settings

#322 DDEle closed 3 months ago
1
XeTLA Zero-Passthrough

#321 DDEle closed 3 months ago
2
Sync ipex(modify prefetch)

#320 sunjiweiswift closed 3 months ago
0
XeTLA enable BF16 tile_t Init

#319 DDEle closed 4 months ago
1
XeTLA Fix Global 1D Store

#318 DDEle closed 4 months ago
1
Xetla GRF Mode Control

#317 DDEle closed 4 months ago
0
XeTLA Sync FMHA Tests

#316 DDEle closed 4 months ago
1
XeTLA Fix load/store 1D

#315 DDEle closed 4 months ago
1
Yi-6B model failed to evaluate

#314 jedcheng closed 4 months ago
1
Int4 dequantize kernel

#313 zhewang1-intc closed 3 months ago
0
sync SYCL code

#312 luoyu-intel closed 3 months ago
0
XeTLA INT4 With BF16 Support

#311 DDEle closed 4 months ago
3
XeTLA Fix Column Major Bugs

#310 DDEle closed 4 months ago
0
Xetla support 2024.2

#309 sunjiweiswift closed 3 months ago
2
BF16 Compute DType on AVX512 ISA

#308 Alavandar08 opened 4 months ago
0
add qwen2 extension test

#307 intellinjun opened 4 months ago
0
[pre-commit.ci] pre-commit autoupdate

#306 pre-commit-ci[bot] opened 4 months ago
0
Xetla support lnl

#305 sunjiweiswift closed 4 months ago
0
XeTLA XMX colmajor

#304 DDEle opened 5 months ago
0
update setuptools

#303 intellinjun closed 5 months ago
0
Enable runtime gpu_arch auto-select based on devices where kernels are executing for gemm_int4 tests; enable device-specific compilation using USE_XETLA (xe_lpg, xe_hpg, xe_hpc).

#302 qgao007 opened 5 months ago
0
[DOC]update README

#301 intellinjun closed 5 months ago
0
update ci

#300 intellinjun closed 5 months ago
0
Mlp fusion

#299 zhewang1-intc closed 4 months ago
2
Init arch Xe2

#298 airMeng opened 5 months ago
0
Add zp no degrad dequant

#297 zhewang1-intc opened 5 months ago
2
[XeTLA] Sync ipex cb5539e

#296 DDEle closed 4 months ago
1
[Bug]fix glm4 convert error

#295 intellinjun closed 5 months ago
0
[Bug]fix glm4 acc error

#294 intellinjun closed 5 months ago
0
[CI]update miniforge

#293 intellinjun closed 5 months ago
0
[BesTLA] Support fp16 for compute_dtype and scale_dtype

#292 luoyu-intel closed 5 months ago
3
[Model]enable glm4-9b

#291 intellinjun closed 5 months ago
2
Whats the different with IPEX-LLM?

#290 manfye opened 5 months ago
0
Bestla Kernels understanding and benchmarking

#289 Alavandar08 opened 5 months ago
8
fix qwen convert error

#288 intellinjun closed 5 months ago
0
developer_document.md need elaboration on determining buffer sizes?

#287 hpcpony opened 5 months ago
1
[Fusion]enable bloom mha fusion

#286 intellinjun closed 5 months ago
1
[Neural Speed] Fix Baichuan, chatGLM1&2&3 acc issue

#285 zhentaoyu closed 5 months ago
3
Performance on Xeon Scalable

#284 regmibijay opened 5 months ago
1
[model]add codestral-22b

#283 intellinjun closed 5 months ago
1
[BesTLA] Add new ISA support: AMX_FP16

#282 luoyu-intel closed 5 months ago
1
[model]Enable qwen2

#281 intellinjun closed 5 months ago
0
[DOC ] Fix cont-batching doc

#280 zhentaoyu closed 5 months ago
0
[BesTLA] Sync compiler's compatibility

#279 luoyu-intel closed 5 months ago
0
[Neural Speed] Fix `ret` when `ignore_prompt`

#278 zhentaoyu closed 5 months ago
2
[CI]fix windows proxy error

#277 intellinjun closed 5 months ago
0