More than 40 operators optimised by agents have been merged into SGLang, with another batch of twenty running for GLM5.2.The count, and which operators it covers, is Humanfia-reported. The numbers for the six below are each pull request's own. Six of them have numbers worth showing.
When each one landed
Swipe the plot sideways to see all of it →
- Merged
Prefill: Kimi Delta Attention on Blackwell
#27488 fixed the CuteDSL prefill kernel for Kimi-Linear, whose real gates overflowed exp() in fp32 and turned the output into NaN, and then made it faster than the Triton path it replaces, by reusing its scratch memory between calls. On B200, with real Kimi-Linear gates:
- CuteDSL prefill
Serving: throughput and time to first token
#22664 turns on FlashInfer's allreduce fusion for Qwen3-Next, where unfused cross-device reduces dominated prefill.
Request throughput · #22664
+71.4%Qwen3-Coder-Next on 4 × H100: 5.49 → 9.41 requests a second.
Mean time to first token
−63.3%456.24 → 167.54 ms. Time per output token fell 49.4%.
#27401 unblocks the TensorRT-LLM NVFP4 fused-MoE kernels for Command-A-Plus on B300. Switch the baseline to compare against vLLM.
#28744 has the router tokenise a prompt once and hand the token ids to the engine, rather than both tokenising it.
- Lower TTFT
Diffusion: denoising and decode
#27524 denoises the early, low-frequency steps at half resolution. The tolerance δ sets how many steps run coarse, so speed rises with it. Drag across the plot to read each model.
Speedup against the tolerance δ
Swipe the plot sideways to see all of it →
#27431 runs the LTX-2 video VAE decode in the channels-last-3d layout, where Conv3d is several times faster on Hopper.
VAE decode stage · #27431
1.41×5.41 → 3.84 s per video; denoising unchanged.
Peak reserved memory
−9.7 GiB71.81 → 62.12 GiB, from fewer transposed copies.
Why an upstream pull request is the test we like most
A contest track is scored by a harness we can read. A production serving framework is scored by maintainers who did not ask for the change, do not care that an agent wrote it, and will reject it if it is slower on a shape we did not think about, harder to maintain, or wrong. Every number above survived that.
Swipe the diagram sideways to see all of it →
1/0 The agent produces candidate operators.