Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>

[Feature] Accelerate VisionAttention by precompute H2D part in every …

[Feature] Accelerate VisionAttention by precompute H2D part in every …

Signed-off-by: Stanley Winata <stanley.winata@amd.com>

[Wave] Add wave extend attention kernel

Signed-off-by: Harsh Menon <harsh@nod-labs.com>

[Wave] Adding logit_cap and layer scaling to API

Also add support for the wave backend to the model
runner. And use Triton decode kernels for now.

[Wave] Run chunked prefill for perf comparison on Wave test

Need to rename the non chunked/regular prefill version because otherwise
rpd will treat it as the same kernel

Signed-off-by: Stanley Winata <stanley.winata@amd.com>

[Wave] Cache the function that loads the wave kernel

Also maintain a global kernel hash to avoid
recomputing the hash on every call.

[Wave] Don't specify block size and enable buffer ops

[Wave] Enable wave runtime and update scheduling API

[Wave] Update API to use wave_compile & WaveCompileOptions

[Wave] Update wave backend and extend attention to latest

[Wave] Add speculative decode kernel

Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com>

cache kernels using lru_cache

Update WaveBackend to use Wave Decode  (sgl-project#6)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

Revert "Update WaveBackend to use Wave Decode  (sgl-project#6)" (sgl-project#7)

This reverts commit eac4599.

Wave Backend decode (sgl-project#8)

* align shapes

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

* fix

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

---------

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

Wave backend fixes (sgl-project#10)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

More fixes to Wave decode (sgl-project#12)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

is_causal

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

Enable the grok in3 model (sgl-project#14)

Set unique cache dir for each worker (sgl-project#16)

update kernel (sgl-project#18)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

updated spec decode test as per wave

Signed-off-by: xintin <gaurav.verma@amd.com>

fix extend (sgl-project#23)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

Refactor paged decode intermediate arrays shapes (sgl-project#24)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

remove dyn symbols (sgl-project#26)

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

cleanup shapes (sgl-project#27)

Some fields were removed from `paged_decode_attention_shape`.

Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>

Remove `mha` param from Wave decode attention kernel (sgl-project#28)

Depends on iree-org/iree-turbine#1039

Signed-off-by: Paul Zhang <paul.zhang@amd.com>

nfc: fix problems reported by linting

update references from iree.turbine to wave_lang

…ate_sgl-jax_discussions

bugfix: use sgl-jax discussions address

fix bug

docs: polish sglang launch & add python request examples

5 adverse-condition tests (4/5 PASS): client disconnect, SIGKILL mid-inference
(with startup preload verification), SIGKILL during write, SIGTERM, abort+save.
New bug #16: SIGTERM graceful shutdown hangs >60s.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: added classification-model.mdx

* fix: modified the description

Replace MTP/EAGLE speculative decoding benchmarks with standard baseline
results using plain sglang serve (no speculative flags), per sgl-cookbook
issue sgl-project#16. All 4 variants (35B-A3B FP8/BF16 and 27B FP8/BF16) benchmarked
on 1× H100 NVL across Chat (1K/1K), Reasoning (1K/8K), and Summarization
(8K/1K) scenarios at concurrency 1, 16, and 64/100.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>