Aan de slagLeaderboardDecode calculatorModellenOrganizationsReportsMatrix runsHardwareBenchmarksMarktplaatsVerhuurProAPI Docs
Taal
Actual Computer — Every computer, one endpoint

Matrix runs

A matrix run measures prompt reading (prefill) and generation (decode) speed in one go, across several context lengths and numbers of simultaneous requests. Run one with lmx speed-test matrix run.

All runs 740

Newest first. Page 22 of 30.
  • Qwen3.5-122B-A10B-FP8H200 NVL 141GB ×2 · vllm 0.26.1rc1.dev638+g52be12cfa · FP86 contexts × 8 concurrency levels
    Peak decode
    5,019.0 tok/s
    Peak prefill
    29.9K tok/s
    by @keennay
  • Qwen3.5-0.8BH200 NVL 141GB · vllm 0.26.1rc1.dev638+g52be12cfa · BF166 contexts × 8 concurrency levels
    Peak decode
    15.6K tok/s
    Peak prefill
    113.3K tok/s
    by @keennay
  • Qwen3.5-0.8BH200 NVL 141GB · vllm 0.26.1rc1.dev638+g52be12cfa · BF166 contexts × 8 concurrency levels
    Peak decode
    17.1K tok/s
    Peak prefill
    132.9K tok/s
    by @keennay
  • Qwen3.5-0.8B-BaseH200 NVL 141GB · vllm 0.26.1rc1.dev638+g52be12cfa · BF166 contexts × 8 concurrency levels
    Peak decode
    11.9K tok/s
    Peak prefill
    118.1K tok/s
    by @keennay
  • Qwen3.5-0.8B-BaseH200 NVL 141GB · vllm 0.26.1rc1.dev638+g52be12cfa · BF166 contexts × 8 concurrency levels
    Peak decode
    17.1K tok/s
    Peak prefill
    122.8K tok/s
    by @keennay
  • INTELLECT-3H200 NVL 141GB ×2 · vllm 0.26.1rc1.dev638+g52be12cfa · BF165 contexts × 8 concurrency levels
    Peak decode
    514.0 tok/s
    Peak prefill
    21.3K tok/s
    by @keennay
  • INTELLECT-3.1H200 NVL 141GB ×2 · vllm 0.26.1rc1.dev638+g52be12cfa · BF165 contexts × 8 concurrency levels
    Peak decode
    516.6 tok/s
    Peak prefill
    20.8K tok/s
    by @keennay
  • INTELLECT-3-FP8H200 NVL 141GB ×2 · vllm 0.26.1rc1.dev638+g52be12cfa · FP85 contexts × 8 concurrency levels
    Peak decode
    1,491.4 tok/s
    Peak prefill
    24.2K tok/s
    by @keennay
  • INTELLECT-2H200 NVL 141GB · vllm 0.26.1rc1.dev638+g52be12cfa · BF164 contexts × 8 concurrency levels
    Peak decode
    625.2 tok/s
    Peak prefill
    8,458 tok/s
    by @keennay
  • Laguna-XS.2H200 NVL 141GB · vllm 0.26.0 · BF166 contexts × 8 concurrency levels
    Peak decode
    3,232.3 tok/s
    Peak prefill
    43.2K tok/s
    by @keennay
  • Laguna-XS.2-NVFP4H200 NVL 141GB · vllm 0.26.0 · NVFP46 contexts × 8 concurrency levels
    Peak decode
    4,818.2 tok/s
    Peak prefill
    32.2K tok/s
    by @keennay
  • Laguna-XS.2-INT4H200 NVL 141GB · vllm 0.26.0 · INT4+INT86 contexts × 8 concurrency levels
    Peak decode
    5,181.7 tok/s
    Peak prefill
    36.2K tok/s
    by @keennay
  • Laguna-XS.2-FP8H200 NVL 141GB · vllm 0.26.0 · FP86 contexts × 8 concurrency levels
    Peak decode
    5,258.4 tok/s
    Peak prefill
    41.2K tok/s
    by @keennay
  • Laguna-XS-2.1H200 NVL 141GB · vllm 0.26.0 · BF166 contexts × 8 concurrency levels
    Peak decode
    3,178.6 tok/s
    Peak prefill
    42.6K tok/s
    by @keennay
  • Laguna-XS-2.1-NVFP4H200 NVL 141GB · vllm 0.26.0 · NVFP46 contexts × 8 concurrency levels
    Peak decode
    4,826.6 tok/s
    Peak prefill
    33.6K tok/s
    by @keennay
  • Laguna-XS-2.1-INT4H200 NVL 141GB · vllm 0.26.0 · INT4+INT86 contexts × 8 concurrency levels
    Peak decode
    5,199.9 tok/s
    Peak prefill
    35.2K tok/s
    by @keennay
  • Laguna-XS-2.1-FP8H200 NVL 141GB · vllm 0.26.0 · FP86 contexts × 8 concurrency levels
    Peak decode
    5,228.0 tok/s
    Peak prefill
    42.4K tok/s
    by @keennay
  • Laguna-XS-2.1H200 NVL 141GB · vllm 0.26.0 · BF166 contexts × 8 concurrency levels
    Peak decode
    1,466.6 tok/s
    Peak prefill
    44K tok/s
    by @keennay
  • Laguna-XS-2.1-NVFP4H200 NVL 141GB · vllm 0.26.0 · NVFP46 contexts × 8 concurrency levels
    Peak decode
    1,618.0 tok/s
    Peak prefill
    33.2K tok/s
    by @keennay
  • Laguna-XS-2.1-INT4H200 NVL 141GB · vllm 0.26.0 · INT4+INT86 contexts × 8 concurrency levels
    Peak decode
    1,695.8 tok/s
    Peak prefill
    34.6K tok/s
    by @keennay
  • Laguna-XS-2.1-FP8H200 NVL 141GB · vllm 0.26.0 · FP86 contexts × 8 concurrency levels
    Peak decode
    1,966.7 tok/s
    Peak prefill
    42.6K tok/s
    by @keennay
  • Laguna-S-2.1-NVFP4H200 NVL 141GB ×2 · vllm 0.27.1 · NVFP48 contexts × 8 concurrency levels
    Peak decode
    3,800.8 tok/s
    Peak prefill
    23.7K tok/s
    by @keennay
  • Laguna-S-2.1-INT4H200 NVL 141GB ×2 · vllm 0.27.1 · INT48 contexts × 8 concurrency levels
    Peak decode
    4,021.1 tok/s
    Peak prefill
    24.2K tok/s
    by @keennay
  • Laguna-S-2.1-FP8H200 NVL 141GB ×2 · vllm 0.27.1 · FP88 contexts × 8 concurrency levels
    Peak decode
    3,595.4 tok/s
    Peak prefill
    29.9K tok/s
    by @keennay
  • Laguna-M.1-NVFP4H200 NVL 141GB ×2 · vllm 0.27.1 · NVFP46 contexts × 8 concurrency levels
    Peak decode
    872.0 tok/s
    Peak prefill
    8,929 tok/s
    by @keennay