Gemini 3.7 Flash
Google. Run in Raw API call.
2 runs on 2 prompts · 0% finished
Results
The best run of each of its 2 tests first, then every other run. Filter them in the explorer
0 finished runs
Lighthouse on a headland at night
Gemini 3.7 Flash
Raw API (one-shot)
raw API baseline
Failed · Did not finish
Numbers: finish rate, cost, wall time, draft scores
- Runs
- 2
- on 2 prompts
- Finished
- 0%
- 0 of 2
- Median cost, finished run
- not recorded
- n=0
- Median wall time, finished run
- not recorded
- n=0
- Recorded spend
- $0
- $0 on unfinished runs; 1 without a recorded cost
From the numbers
Set against the other models with at least 3 runs (top or bottom third), and its draft-score ranks within single prompts. Nothing here is averaged across prompts.
Strengths
- Nothing stands out with enough runs to say so.
Weaknesses
- 1 run ended as the model's own failure (DNF, class MODEL)
- Other unfinished runs, not counted against the model: 1 INFRA
Draft scores by prompt
No scored result yet (only prompts with a published rubric are scored).
By harness
| Raw API call | 2 | 0 | 0% | not recorded: no finished run with a recorded cost | not recorded: no finished run with a recorded time |
|---|
All 2 tests and experiment collections
Experiment collections: SVG (2) · Free lane (1)