-
Notifications
You must be signed in to change notification settings - Fork 43
Pull requests: flashinfer-ai/flashinfer-bench
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add setup-aware kernel-only timing for benchmark workloads
#427
opened Jul 1, 2026 by
MengYu10151
Loading…
feat(scripts): sanitize FlashInfer level-3 logs into workload JSONL
#425
opened May 1, 2026 by
yyihuang
Contributor
Loading…
4 tasks done
fix: guard against ZeroDivisionError when computing speedup_factor
#416
opened Apr 23, 2026 by
factnn
Loading…
fix: handle None expected shape for scalar inputs in load_safetensors
#415
opened Apr 23, 2026 by
factnn
Loading…
fix: randomize float inputs each iteration to prevent output caching
#413
opened Apr 23, 2026 by
factnn
Loading…
fix: remove extra def_name subdirectory in workload blob path
#412
opened Apr 23, 2026 by
factnn
Loading…
feat: onboard Kimi K2.6 for mla_paged_decode_h8_ckv512_kpe64_ps1
#410
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add top_k_sampling_from_probs_v163840 definition (Kimi K2.6)
#409
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add gemm_n3072_k8192 definition
#408
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add mla_paged_decode_h8_ckv512_kpe64_ps1 definition (Kimi K2.6)
#407
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add model:kimi-k2.6 coverage for mla_paged_decode_h8_ckv512_kpe64_ps1
#404
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add gemm_n5120_k3072 and gemm_n3072_k3072 definitions (Llama 3.2 3B)
#403
opened Apr 21, 2026 by
flashinfer-bot
Contributor
Loading…
feat: add Llama 3.2 1B model entry (coverage + sglang configs + web catalog)
#402
opened Apr 20, 2026 by
ksgr5566
Loading…
feat: integrate Kimi K2.5 (model coverage + serving configs; partial, MoE blocked)
#401
opened Apr 18, 2026 by
ksgr5566
Loading…
feat: add fused_add_rmsnorm_h6144 definition, workload, and coverage
#400
opened Apr 16, 2026 by
ksgr5566
Loading…
feat: add rmsnorm_h6144 definition, workloads, and reference test
#399
opened Apr 16, 2026 by
000FLMS
Loading…
feat: add trtllm_fp8_block_scale_moe_topk8_e256_h3072_i1536 definition (MiniMax M2)
#397
opened Apr 14, 2026 by
yyihuang
Contributor
Loading…
2 tasks
fix: improve sampling evaluator for large-nucleus distributions
#396
opened Apr 14, 2026 by
yyihuang
Contributor
Loading…
2 tasks
feat: add gqa_paged_decode_h5_kv1_d128_ps64 definition (Llama 4 Scout/Maverick, TP=8)
#389
opened Apr 13, 2026 by
yyihuang
Contributor
Loading…
2 tasks
feat: add moe_fp8_block_scale_ds_routing_topk8_ng1_kg1_e256_h3072_i1536 definition
#377
opened Apr 12, 2026 by
yyihuang
Contributor
Loading…
3 tasks done
feat: add gemm_n3072_k8192 definition and reference test
#375
opened Apr 12, 2026 by
yyihuang
Contributor
Loading…
feat: add gemm_n16384_k3072 definition and reference test
#374
opened Apr 12, 2026 by
yyihuang
Contributor
Loading…
feat: add top_k_top_p_sampling_from_probs_v200064 reference test (MiniMax M2)
#373
opened Apr 12, 2026 by
yyihuang
Contributor
Loading…
feat: add top_p_sampling_from_probs_v200064 reference test (MiniMax M2)
#372
opened Apr 12, 2026 by
yyihuang
Contributor
Loading…
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.