yy-fighting 7d2ed7f05d Add vLLM+MTP speculative decoding benchmark vs default on H200 ...

- Experiment: experiments/dsv4_h200_vllm_mtp_vs_default/
- Use official DeepSeek-V4 recipe: --speculative-config '{method: mtp, num_speculative_tokens: 1}'
- Reuse default-side results from dsv4_h200_vllm_dspark_vs_default
- Add TTFT analysis for DSpark vs default trade-offs

2026-07-08 16:13:01 +00:00

..

dsv4_h200_64k_sglang_vs_vllm

docs: move SLO_STANDARDS.md from scripts/ to docs/; update references

2026-07-08 11:06:49 +00:00

dsv4_h200_256k_4k_probe

docs: move SLO_STANDARDS.md from scripts/ to docs/; update references

2026-07-08 11:06:49 +00:00

dsv4_h200_dspark

feat: add H200 SGLang baseline experiment and fix health check

2026-07-08 06:56:45 +00:00

dsv4_h200_long_context_matrix

results(long-context-matrix): complete 64k/128k/256k matrix run 20260708-100016; use common scripts

2026-07-08 11:21:15 +00:00

dsv4_h200_max_context_length

results(max-context): complete H200 max context length exploration run 20260708-143534

2026-07-08 15:33:48 +00:00

dsv4_h200_sglang

feat: add H200 SGLang baseline experiment and fix health check

2026-07-08 06:56:45 +00:00

dsv4_h200_sglang_vs_vllm

docs: move SLO_STANDARDS.md from scripts/ to docs/; update references

2026-07-08 11:06:49 +00:00

dsv4_h200_vllm

fix(dsv4_h200_vllm): scenario array parsing and sglang output format

2026-07-08 06:31:03 +00:00

dsv4_h200_vllm_dspark_vs_default

Add vLLM+MTP speculative decoding benchmark vs default on H200

2026-07-08 16:13:01 +00:00

dsv4_h200_vllm_mtp_vs_default

Add vLLM+MTP speculative decoding benchmark vs default on H200

2026-07-08 16:13:01 +00:00

dsv4_p800_sglang

fix(p800): managed-mode server startup and smoke-test defaults

2026-07-08 05:49:55 +00:00

legacy_bench_results

refactor: move final legacy result dspark_grid into experiments/legacy_bench_results/

2026-07-08 06:40:39 +00:00

TEMPLATE

refactor(common): centralize parse/compare/warmup; add TEMPLATE and EXPERIMENT_GUIDE

2026-07-08 10:48:43 +00:00