GitLab

Co-authored-by: Varun Sundar Rabindranath <varunsundar08@gmail.com>
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com>

Switching from torch._scaled_mm to vLLM's cutlass fp8 kernels when supported as we are seeing 5-15% improvement in e2e performance on neuralmagic/Meta-Llama-3-8B-Instruct-FP8

see https://docs.google.com/spreadsheets/d/1GiAnmzyGHgZ6zL_LDSTm35Bdrt4A8AaFEurDlISYYA4/ for some quick e2e benchmarks and #5144 for comparisons across different GEMM sizes.

This PR enables the fused topk_softmax kernel used in moe layer for HIP

This PR adds Triton kernel configs for the MoE kernel for MI300X

Co-authored-by: Elisei Smirnov <el.smirnov@innopolis.university>

Co-authored-by: Varun Sundar Rabindranath <varunsundar08@gmail.com>
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com>

The 2nd PR for #4532.

This PR supports loading FP8 kv-cache scaling factors from a FP8 checkpoint (with .kv_scale parameter).

Currently we need to call rotary embedding kernel for each LoRA, which makes it hard to serve multiple long context length LoRA. Add batched rotary embedding kernel and pipe it through.

It replaces the rotary embedding layer to the one that is aware of multiple cos-sin-cache per scaling factors.

Follow up of https://github.com/vllm-project/vllm/pull/3095/files

Co-authored-by: Robert Shaw <rshaw@neuralmagic.com>

[Core][2/N] Model runner refactoring part 2. Combine prepare prefill / decode to a single API (#4681)

This PR combines prepare_prompt and prepare_decode into a single API. This PR also coelsce the attn metadata for prefill/decode to a single class and allow to slice them when running attn backend.

It also refactors subquery_start_loc which was not refactored in the previous PR

This PR improves the FP8 performance of linear layers, which had been lacking before (#4118 (comment) and #4118 (comment)).

We noticed that CUBLASLt can find a better algorithm if the first dimension of the matrix is greater than 16. So this PR enlarges matrices appropriately during quantization. This improves FP8 performance and removes the performance regression vs. FP16, in many cases exceeding FP16 performance.

Here are benchmarks on llama3 70b (ITL numbers for 1000 input and 50 output tokens at fixed qps and at TP 4), all FP8 measurements are for dynamic quantization:

qps = 1: 24 ms (FP8, this PR), 32 ms (FP8, previous main), 26 ms (FP16)
qps = 2: 26 ms (FP8, this PR), 34ms (FP8, previous main), 28 ms (FP16) 
qps = 4: 33 ms (FP8, this PR), 44 ms (FP8, previous main), 36 ms (FP16)
qps = 6: 46 ms (FP8, this PR), 56 ms (FP8, previous main), 54 ms (FP16)
qps = 8: 85 ms (FP8, this PR), 85 ms (FP8, previous main), 138 ms (FP16)

Co-authored-by: Dash Desai <1723932+iamontheinet@users.noreply.github.com>
Co-authored-by: Aurick Qiao <qiao@aurick.net>
Co-authored-by: Aurick Qiao <aurick.qiao@snowflake.com>
Co-authored-by: Aurick Qiao <aurickq@users.noreply.github.com>
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>

Co-authored-by: Cade Daniel <edacih@gmail.com>