roikoren755 b021332339 [NemotronH] Add latent MoE support (#16227) ...

Signed-off-by: Roi Koren <roik@nvidia.com>

2026-01-02 22:08:58 +08:00

..

bench_attention_sink

…

bench_in_batch_prefix

…

benchmark_batch

…

benchmark_vllm_060

…

blog_v0_2

…

boolq

…

ceval

…

deepseek_v3

update benchmark README to use --fp8-gemm-backend instead of env var (#15689)

2025-12-23 23:23:31 -08:00

dspy

…

generative_agents

…

gpt_oss

…

gsm8k

…

hellaswag

…

hf3fs

[hotfix]: Add missing args for 3FS bench_client.py (#14791)

2025-12-14 11:13:19 -10:00

hicache

Update benchmarks to use HF token from environment. (#15421)

2025-12-18 13:47:27 -08:00

json_decode_regex

…

json_jump_forward

…

json_schema

…

kernels

[NemotronH] Add latent MoE support (#16227)

2026-01-02 22:08:58 +08:00

line_retrieval

…

llava_bench

…

llm_judge

…

long_json_decode

…

lora

Fix lora test (#13479)

2025-11-18 12:21:57 +08:00

mmlu

…

mmmu

ministral3 (#14251)

2025-12-04 14:31:26 -08:00

mtbench

Fix Illegal Memory Access when fa3 + spec + topk + page_size > 1 (#15469)

2025-12-24 00:13:57 +08:00

multi_chain_reasoning

…

multi_document_qa

…

multi_turn_chat

Tiny move files to utils folder (#11166)

2025-10-03 22:40:06 +08:00

prefill_only

Add overlap scheduling for embeddings code path (#14032)

2025-12-24 18:24:18 -08:00

react

…

reasoning_benchmark

…

tip_suggestion

…

tree_of_thought_deep

…

tree_of_thought_v0

…