2 Commits

Author	SHA1	Message	Date
yy-fighting	1482601ae5	Add num_prompts = 5×concurrency convention and update all experiments ... - Document the rule in docs/EXPERIMENT_GUIDE.md section 7 - Update all config.env SCENARIOS to follow the 5x rule - Migrate dsv4_h200_sglang, dsv4_h200_vllm, dsv4_p800_sglang from global NUM_PROMPTS to per-scenario num_prompts - Keep long-context phase2 as exception (low counts documented) - Add legacy 3-field fallback in run_bench.sh loops	2026-07-09 09:26:41 +00:00
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