Created
January 27, 2026 03:34
-
-
Save sammcj/e45c2ad08a4191f0fbbaa1b842a5a778 to your computer and use it in GitHub Desktop.
vLLM fused_moe tuned configuration RTX3090
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "triton_version": "3.5.1", | |
| "1": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 32, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "2": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "4": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 32, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "8": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 32, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "16": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "24": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "32": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 32, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "48": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 32, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "64": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "96": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "128": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "256": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "512": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "1024": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "1536": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "2048": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "3072": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| }, | |
| "4096": { | |
| "BLOCK_SIZE_M": 64, | |
| "BLOCK_SIZE_N": 128, | |
| "BLOCK_SIZE_K": 128, | |
| "GROUP_SIZE_M": 16, | |
| "num_warps": 8, | |
| "num_stages": 3 | |
| } | |
| } |
Author
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Created using https://github.com/MissionSquad/vllm-moe-configs/pull/1/changes#diff-df577f927c5560caca03b81a8eff9390de41552b0f26391c72ebc6905dea8a5c