Modell-Vergleich
AGGREGAT-STATS ÜBER ALLE LÄUFE
Pipelines
| Pipeline | Runs | Fehler | Champion% | Gut% (User) | Avg-Latenz | Avg-€ | Tokens (in/out) |
|---|---|---|---|---|---|---|---|
Production (Claude Sonnet 4) production-aktuell | 4 | 0 | 18.2% | 0.0% | — | $0.0062 | 3352 / 980 |
Qwen3.5 122B (Mittwald) mittwald-qwen-3-5-122b | 4 | 0 | 9.1% | 0.0% | 17.7s | $0.00 | 2470 / 9361 |
Qwen3.6 35B (Mittwald) mittwald-qwen-3-6-35b | 3 | 0 | 9.1% | 0.0% | 10.0s | $0.00 | 1848 / 5936 |
Opus-Judge (Auto-Bewertung) judge-opus-4-7 | 4 | 0 | 0.0% | 0.0% | 6.8s | $0.0793 | 12772 / 1677 |
Production (Claude Sonnet 4)
production-aktuell
- Runs
- 4
- Fehler
- 0
- Champion%
- 18.2%
- Gut% (User)
- 0.0%
- Avg-Latenz
- —
- Avg-€
- $0.0062
- Tokens (in/out)
- 3352 / 980
Qwen3.5 122B (Mittwald)
mittwald-qwen-3-5-122b
- Runs
- 4
- Fehler
- 0
- Champion%
- 9.1%
- Gut% (User)
- 0.0%
- Avg-Latenz
- 17.7s
- Avg-€
- $0.00
- Tokens (in/out)
- 2470 / 9361
Qwen3.6 35B (Mittwald)
mittwald-qwen-3-6-35b
- Runs
- 3
- Fehler
- 0
- Champion%
- 9.1%
- Gut% (User)
- 0.0%
- Avg-Latenz
- 10.0s
- Avg-€
- $0.00
- Tokens (in/out)
- 1848 / 5936
Opus-Judge (Auto-Bewertung)
judge-opus-4-7
- Runs
- 4
- Fehler
- 0
- Champion%
- 0.0%
- Gut% (User)
- 0.0%
- Avg-Latenz
- 6.8s
- Avg-€
- $0.0793
- Tokens (in/out)
- 12772 / 1677