Decision model
Fastino GLiDE
1 question per request
67.9% of the 2,198 questions right (65.9% to 69.8%, 95% interval). Back to the leaderboard.
How we called it
- Kind
- Decision model
- Access
- Fastino System One API
- Paid as
- Provider service
- Model ID
fastino/GLiDE- Note
- Hosted decision model; Fastino describes it as reasoning on uncertain decisions
- Confidence from
- Option score
- Run
- 129, 6 October 2026
Settings
- Reasoning
- off
- Request format
- 1 question per request
- Requests at once
- 4
- Temperature
- not recorded
- Harness seed
- 0
- Harness commit
- not recorded
- Prompt version
- not recorded
Results
| Questions | Accuracy | 95% interval | Right |
|---|---|---|---|
| All questions | 67.9% | 65.9% to 69.8% | 1,492 of 2,198 |
| Reading | 87.8% | 85.1% to 90.1% | 576 of 656 |
| Grammar | 59.4% | 56.9% to 61.8% | 916 of 1,542 |
Comparison group: pqr. Compare with another setup.
Accuracy by topic
Confidence
- Answers kept at 99% or more
- 121, of which 0 wrong (share right: 96.9% to 100.0%, 95% interval); 2,077 skipped, the model less than 99% sure
- Calibration gap
- 4.5 points
- AUROC, all questions
- 0.81 0.79 to 0.83
- AUROC, reading
- 0.91 0.88 to 0.93
- AUROC, grammar
- 0.74 0.72 to 0.76
- Brier score
- 0.165
- Accuracy of its surest answers
- most confident fifth: 99.5% right; most confident half: 89.1% right; most confident four fifths: 75.9% right
- Typical score of a right answer
- 76.7%
- Typical score of a wrong answer
- 53.2%
Repeats
We did not repeat this setup.
Latency
- Latency, median request
- 2.09 s (this run)
- Latency, 90th percentile
- 2.28 s
- Requests at once
- 4
- Throughput
- 152.1 questions per minute of request time
- Request time
- 14.4 minutes, over the requests whose answers were kept
Latency: the time from sending one request to receiving its answer, measured from our computer, network and any retries within the request included. Request time: the time at least one request was in flight, over the requests whose answers were kept. It is not the run's elapsed time: it leaves out requests that failed and any pause before a run was resumed.
Cost
- $ per 1,000 questions
- $0.16
- Whole run
- $0.35
- Basis
- Estimated from published prices and recorded token usage
- Price
- $0.15 input, output free per 1M tokens (source, 2026-10-08)
- Input tokens per question
- 1,067
- Output tokens per question
- 254
Going first in front of an LLM
Replay of recorded answers, cutoff chosen on half the questions, averaged over the splits. Savings are estimates from API prices; the Method page explains how.
| LLM behind it | Share taken | Accuracy change | Saving |
|---|---|---|---|
| Gemini 3.8 Flashhigh reasoning · 16 questions per request | 20.4% | −0.10 points | 14% |
| GPT-6 Astrahigh reasoning · 16 questions per request | 20.1% | −0.09 points | 17% |
| GPT-6.1 Solhigh reasoning · 16 questions per request | 20.9% | −0.10 points | 11% |
| GPT-6 Astralow reasoning · 16 questions per request | 20.1% | −0.09 points | 17% |
| GPT-6 Astralow reasoning · 1 question per request | 20.1% | −0.09 points | 20% |
| Gemini 3.8 Flashlow reasoning · 1 question per request | 22.2% | −0.18 points | 6% |
| GPT-6.1 Sollow reasoning · 1 question per request | 21.9% | −0.14 points | 19% |
| GPT-6.1 Sollow reasoning · 16 questions per request | 20.2% | −0.10 points | 6% |
| Claude Opus 5.5high reasoning · 16 questions per request | 20.7% | −0.09 points | 18% |
| GPT-6.1 Terrahigh reasoning · 16 questions per request | 20.4% | −0.10 points | -11% |
| Claude Opus 5.5low reasoning · 1 question per request | 21.6% | −0.14 points | 20% |
| Gemini 3.8 Flashlow reasoning · 16 questions per request | 21.3% | −0.13 points | -38% |
| GPT-6.1 Terralow reasoning · 16 questions per request | 23.3% | −0.15 points | -14% |
| Claude Opus 5.5low reasoning · 16 questions per request | 21.8% | −0.10 points | 19% |
| Claude Sonnet 5.5high reasoning · 1 question per request | 24.7% | −0.15 points | 23% |
| Claude Sonnet 5.5high reasoning · 16 questions per request | 44.6% | −0.20 points | 40% |