Search results

cpu offloading doesn't work

Status: Open.

#556 In intel/llm-scaler;

· competitiveNN opened on Jul 22, 2026

[vllm] Pre-quantized FP8 Qwen3.6-27B-FP8 fails on XPU: missing PlatformEnum.XPU in FP8 block kernel table (0.21.0-b1); AttributeError at construction (0.14.0-b8.3.2)

Status: Open.

#550 In intel/llm-scaler;

· MRdodoe opened on Jul 19, 2026

[omni_xpu_kernel] Lumina/Z-Image / SDXL 模型在 Arc B580 上测试报告（oneAPI 2026.1）

Status: Open.

#548 In intel/llm-scaler;

· allanmeng opened on Jul 16, 2026

AttributeError: 'RowParallelLinear' object has no attribute 'weight' — GDN linear attention ESIMD path fails on GPTQ-quantized hybrid models (Qwen3.5-MoE)

Status: Open.

#533 In intel/llm-scaler;

· doopeworld opened on Jul 11, 2026

Intel llm-scaler vLLM crashes on Gemma 4 due to k_descale stride assertion

Status: Open.

#532 In intel/llm-scaler;

· knikkinen opened on Jul 11, 2026

[Bug] OpenAI-compatible streaming tool_calls omit function.name after first chunk, breaking OpenCode with Qwen3.6-27B on llm-scaler-vllm 0.14.0-b8.3.2

Status: Open.

#516 In intel/llm-scaler;

· mikellantalde opened on Jul 7, 2026

Add a xpu int8_gemm backend kernel for comfy-kitchen

Status: Open.

#512 In intel/llm-scaler;

· matalama80td3l opened on Jul 6, 2026

Qwen/Qwen3.6-35B-A3B BF16 inference is failing on Intel Arc Pro B70 cards

Status: Open.

#501 In intel/llm-scaler;

· mayavijx opened on Jun 29, 2026

tool_choice: "required" crashes EngineCore with SYCL device exception; "auto" works fine with identical request

Status: Open.

#493 In intel/llm-scaler;

· diablo581 opened on Jun 24, 2026

Bug Report: Pipeline-Parallel (PP=2) Engine Crash / Hang on Dual Battlemage GPUs (Arc Pro B70 + Arc B580)

Status: Open.

#489 In intel/llm-scaler;

· doopeworld opened on Jun 18, 2026

PyTorch 2.x runtime error

Status: Open.

#481 In intel/llm-scaler;

· zhou80bin opened on Jun 15, 2026

[Bug]: INT4 GPTQ MoE model crashes with UR_RESULT_ERROR_OUT_OF_RESOURCES on Intel Arc Pro B70 — dense models work fine

Status: Open.

#479 In intel/llm-scaler;

· doopeworld opened on Jun 13, 2026