anthropic ·claude-sonnet-4-5May 18, 06:01 PM
Asset 88Lb01q
Score
0.9
Latency
7.07s
Cost
$0.0151
Workflow Eval Detail
Ask Questions
Answers natural-language questions about a video by retrieving relevant context and answering with a concise response.
Latest Runcompleted
muxinc/ai
main·c15880c·@mux/ai v0.22.0
Cases
7
Avg Score
0.95
Avg Latency
5.36s
Avg Cost
$0.0036
Avg Cost / Min
$0.0063/min
Avg Tokens
3,694
TL;DR
On this small 7-case suite, all providers are accurate, but Google flash/flash-lite models provide the best overall tradeoff of quality, latency, and cost for Ask Questions.
Best Quality
google
gemini-2.5-flash
Fastest
google
gemini-3.1-flash-lite
Most Economical
google
gemini-3.1-flash-lite
What we measure
Each eval run captures efficacy, efficiency, and expense. We use this data to compare providers and track regressions over time.
Efficacy
Quality + correctness
Efficiency
Latency + token usage
Expense
Cost per request
Workflow snapshot
Suite statussuccess
Suite average score0.95
Suite duration37.54s
Last suite runMay 18, 05:59 PM
Evaluation criteria
From eval testsWe score answer accuracy and response integrity while tracking latency, token usage, and cost.
Question set3 checks
Question
Has on-screen text?
Answer: YesConf0.94
Question
Is the speaker visible?
Answer: NoConf0.88
Question
Scene indoors?
Answer: YesConf0.91
Accuracy
Match expected yes/no outputs.
Format
Required fields + response shape.
Integrity
Confidence and reasoning present.
Throughput
Latency and cost stay within targets.
Response validation
Efficacy checks
- Answers match expected yes/no outputs.
- Response includes required fields and answer structure.
- Confidence is 0-1 and reasoning strings are non-empty.
- Response preserves asset ID and storyboard URL.
Efficiency targets
- Latency: scores are normalized between 0 and 1. Under 8s earns 1.0; past 12s trends toward 0.
- Token usage: scores are normalized between 0 and 1. Under 2,900 tokens earns 1.0; higher usage reduces the score.
- Usage data must include total tokens for cost analysis.
Expense guardrails
- Estimated cost under $0.012 per request for full score.
Provider breakdown
Run c15880cEfficacy scoreHigher is better
LatencyLower is better
Token UsageLower is better
CostLower is better
| Provider | Model | Cases | Avg Score | Avg Latency | Avg Tokens | Avg Cost | Avg Cost / Min |
|---|---|---|---|---|---|---|---|
| anthropic | claude-sonnet-4-5 | 1 | 0.9 | 7.07s | 4,451 | $0.0151 | $0.0263/min |
| gemini-2.5-flash | 1 | 1 | 3.73s | 2,881 | $0.0017 | $0.0029/min | |
| gemini-3-flash-preview | 1 | 0.96 | 4.61s | 3,796 | $0.0022 | $0.0039/min | |
| gemini-3.1-flash-lite | 1 | 0.98 | 2.58s | 3,480 | $0.0011 | $0.0018/min | |
| gemini-3.1-flash-lite-preview | 1 | 0.97 | 2.67s | 3,489 | $0.0011 | $0.0019/min | |
| openai | gpt-5-mini | 1 | 0.85 | 13.4s | 4,750 | $0.0023 | $0.004/min |
| openai | gpt-5.1 | 1 | 1 | 3.48s | 3,013 | $0.0018 | $0.0032/min |
Recent cases
Latest 6google ·gemini-2.5-flashMay 18, 06:01 PM
Asset 88Lb01q
Score
1
Latency
3.73s
Cost
$0.0017
google ·gemini-3-flash-previewMay 18, 06:01 PM
Asset 88Lb01q
Score
0.96
Latency
4.61s
Cost
$0.0022
google ·gemini-3.1-flash-liteMay 18, 06:01 PM
Asset 88Lb01q
Score
0.98
Latency
2.58s
Cost
$0.0011
google ·gemini-3.1-flash-lite-previewMay 18, 06:01 PM
Asset 88Lb01q
Score
0.97
Latency
2.67s
Cost
$0.0011
openai ·gpt-5-miniMay 18, 06:01 PM
Asset 88Lb01q
Score
0.85
Latency
13.4s
Cost
$0.0023