fm-bench 0.4.0 → 0.4.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +1 -1
- package/package.json +1 -1
- package/src/table.js +4 -2
package/README.md
CHANGED
|
@@ -160,7 +160,7 @@ Table output uses semantic ANSI color on interactive terminals:
|
|
|
160
160
|
- yellow: marginal, partial, or near a budget.
|
|
161
161
|
- red: failing a budget, unstable, or slower/lower than peers.
|
|
162
162
|
|
|
163
|
-
Success rate, goodput, repeatability, and CV use fixed benchmark thresholds. Throughput columns use relative ranking within the current run because “good” depends on the machine, model, prompt mix, and concurrency. TTFT, E2E, and TPOT use SLO thresholds when you pass `--slo-ttft-ms`, `--slo-e2e-ms`, or `--slo-tpot-ms`; otherwise they use lower-is-better relative ranking across the models and operating points in the report.
|
|
163
|
+
Success rate, goodput, repeatability, and CV use fixed benchmark thresholds. CV is green at `<=10%`, yellow at `<=25%`, and red above `25%` because higher CV means less steady latency. Throughput columns use relative ranking within the current run because “good” depends on the machine, model, prompt mix, and concurrency. TTFT, E2E, and TPOT use SLO thresholds when you pass `--slo-ttft-ms`, `--slo-e2e-ms`, or `--slo-tpot-ms`; otherwise they use lower-is-better relative ranking across the models and operating points in the report.
|
|
164
164
|
|
|
165
165
|
Use `--color` to force ANSI colors in captured logs, or `--no-color` for plain output. `NO_COLOR=1` disables automatic color and `FORCE_COLOR=1` enables it.
|
|
166
166
|
|
package/package.json
CHANGED
package/src/table.js
CHANGED
|
@@ -251,8 +251,10 @@ function terminalWidth(options = {}) {
|
|
|
251
251
|
}
|
|
252
252
|
|
|
253
253
|
function compactLegend(width) {
|
|
254
|
-
|
|
255
|
-
|
|
254
|
+
return [
|
|
255
|
+
'TTFT = first streamed output. E2E = full response. TPOT = post-first-token decode cadence.',
|
|
256
|
+
'CV = latency variation; lower is steadier: green <=10%, yellow <=25%, red >25%.'
|
|
257
|
+
].map((line) => truncate(line, width)).join('\n');
|
|
256
258
|
}
|
|
257
259
|
|
|
258
260
|
function formatSlo(slo = {}) {
|