Holo3 122B A10B
H Company · holo3-122b-a10b
- Latency
- 1.38 s
- Cost
- $0.00055
- Attempts
- 1
{"found":true}
Element presence scenario · 2026-08-13
Is the Restart device button present in IoT control center?
Determine whether the Restart device button is visibly present.
Expected answerPresent
1440 × 900 · 9 model configurations
Exact input
Prompt
You are evaluating a user interface screenshot for test automation.
Determine whether the single UI element described below is visibly present. Return found=true if it is present and found=false if it is absent.
Target: the Restart device button
Return only data matching the supplied JSON schema. Do not describe your reasoning.
This dated pilot used the geometry-capable prompt shown above. The published presence result uses only the returned found value; geometry is neither scored nor displayed.
Ground truth
Model evidence
9 model results
Presence pages report only the boolean classification. Original model output remains visible for auditability.
Correct
Grok 4.5
xAI · x-ai/grok-4.5- Latency
- 7.43 s
- Cost
- $0.0032
- Attempts
- 1
{"found": true}
GPT-5.4
OpenAI · openai/gpt-5.4- Latency
- 4.81 s
- Cost
- $0.0044
- Attempts
- 1
{"found":true}
Gemini 3.6 Flash
Google AI Studio · google/gemini-3.6-flash- Latency
- 9.20 s
- Cost
- $0.0022
- Attempts
- 1
{"found": true}
GPT-5.6 Sol
OpenAI · openai/gpt-5.6-sol- Latency
- 5.41 s
- Cost
- $0.0110
- Attempts
- 1
{"found":true}
GPT-5.6 Luna
OpenAI · openai/gpt-5.6-luna- Latency
- 8.63 s
- Cost
- $0.00022
- Attempts
- 1
{"found":true}
Claude Sonnet 5
Anthropic · anthropic/claude-sonnet-5- Latency
- 10.46 s
- Cost
- $0.0043
- Attempts
- 1
{"found":true}
Claude Opus 5
Anthropic · anthropic/claude-opus-5- Latency
- 8.36 s
- Cost
- $0.0107
- Attempts
- 1
{"found":true}
Gemini 3.1 Pro Preview
Google AI Studio · google/gemini-3.1-pro-preview- Latency
- 26.31 s
- Cost
- $0.0024
- Attempts
- 1
{"found": true}