Bob Fleet Metrics
+Last 30 days of autonomous agent sessions — generated 2026-07-27 02:59 UTC
+diff --git a/fleet-metrics/index.html b/fleet-metrics/index.html new file mode 100644 index 000000000..485581b0d --- /dev/null +++ b/fleet-metrics/index.html @@ -0,0 +1,261 @@ + + +
+ + +Last 30 days of autonomous agent sessions — generated 2026-07-27 02:59 UTC
+Daily autonomous sessions with productive overlay and average quality score trend
+ + + ++ Productive rate and LLM judge quality score per model+harness combination + (min 8 sessions, last 30 days). + ■ claude-code + ■ codex + ■ gptme +
+ +| Harness:Model | +Productive rate | +% | +Score | +Sessions | +
|---|---|---|---|---|
| claude-code:Sonnet 4.6 | ++ + | +87.8% | +0.55 | +4,237 | +
| claude-code:Haiku 4.5 | ++ + | +95.5% | +— | +797 | +
| claude-code:Sonnet (old) | ++ + | +52.2% | +0.57 | +722 | +
| codex:GPT-5.5 | ++ + | +98.8% | +0.49 | +521 | +
| gptme:GPT-5.6 (sub) | ++ + | +54.2% | +— | +330 | +
| claude-code:Opus | ++ + | +94.1% | +0.42 | +256 | +
| codex:GPT-5.4 | ++ + | +96.6% | +0.51 | +234 | +
| gptme:GPT-5.6 | ++ + | +93.6% | +0.53 | +233 | +
| gptme:DeepSeek V4 | ++ + | +85.4% | +— | +233 | +
| gptme:DeepSeek V4 | ++ + | +74.3% | +0.46 | +101 | +
| claude-code:Haiku (old) | ++ + | +67.0% | +0.69 | +97 | +
| claude-code:Fable 5 | ++ + | +98.8% | +0.72 | +85 | +