| 370 | opt(fp, c, "--kv-disk-dir DIR", "Enable disk KV checkpoints in DIR."); |
| 371 | opt(fp, c, "--kv-disk-space-mb N", "Disk budget. Default when enabled: 4096"); |
| 372 | opt(fp, c, "--kv-cache-min-tokens N", "Do not save/load checkpoints shorter than N. Default: 512"); |
| 373 | opt(fp, c, "--kv-cache-cold-max-tokens N", "Save cold first prompts up to N tokens. 0 disables. Default: 30000"); |
| 374 | opt(fp, c, "--kv-cache-continued-interval-tokens N", "Save aligned continued frontiers. 0 disables. Default: 10000"); |
| 375 | opt(fp, c, "--kv-cache-boundary-trim-tokens N", "Trim tail tokens for cold boundary saves. Default: 32"); |
| 376 | opt(fp, c, "--kv-cache-boundary-align-tokens N", "Align cold boundary saves to this multiple. Default: 2048"); |
| 377 | opt(fp, c, "--kv-cache-reject-different-quant", "Reject checkpoints written with different routed-expert quantization."); |
| 378 | opt(fp, c, "--disable-exact-dsml-tool-replay", "Disable exact sampled DSML tool replay map."); |
| 379 | opt(fp, c, "--tool-memory-max-ids N", "Exact tool-call IDs kept in RAM. Default: 100000"); |
| 380 | fputc('\n', fp); |
| 381 | } |
| 382 | |
| 383 | static void print_bench_specific(FILE *fp, const help_colors *c) { |
| 384 | title(fp, c, "Benchmark Input"); |
| 385 | opt(fp, c, "--prompt-file FILE", "Raw benchmark text; token sequence is sliced at each frontier."); |
| 386 | opt(fp, c, "--chat-prompt-file FILE", "Render FILE as one no-thinking chat user message."); |
| 387 | opt(fp, c, "-sys, --system TEXT", "System prompt used only with --chat-prompt-file."); |
| 388 | fputc('\n', fp); |
| 389 | title(fp, c, "Benchmark Sweep"); |
| 390 | opt(fp, c, "--ctx-start N", "First measured frontier. Default: 2048"); |
| 391 | opt(fp, c, "--ctx-max N", "Last measured frontier. Default: 32768"); |
| 392 | opt(fp, c, "--ctx-alloc N", "Allocated context. Default: ctx-max + gen-tokens + 1"); |
| 393 | opt(fp, c, "--step-mul F", "Multiplicative step. Default: 1"); |
| 394 | opt(fp, c, "--step-incr N", "Linear step when --step-mul is 1. Default: 2048"); |
| 395 | opt(fp, c, "--gen-tokens N", "Greedy decode tokens per frontier. 0 for pure prefill. Default: 128"); |
| 396 | opt(fp, c, "--teacher-forced-decode", "Decode the following prompt tokens instead of each predicted argmax."); |
no test coverage detected