Model

Llama 3.2 90B (Vision)

Test

Data extraction

Scenario

Indirect birth year

Run ID

2024-09-26-01J8Q0QADC4FESPMXSA13JFWNP

Model input

User/Human
What year was Sarah born? Only return the year as a four-digit number. In 2023, Sarah celebrated her 30th birthday with a big party. Her younger sister, who is 5 years younger, couldn't attend because she was studying abroad.

Model response

AI
2023 - 30 = 1993
id"gen-kqrr65GIIS9Z7gJgzxBozo9xAyvf" 
total_cost0.0000657
upstream_id"d855c3e4-6e84-4436-b39c-35ad5958fd78" 
created_at"2024-09-26T11:16:09.978869+00:00" Thursday, September 26th, 2024 at 11:16:09 AM GMT+00:00
model"meta-llama/llama-3.2-90b-vision-instruct" 
app_id182717
streamedtrue
cancelledfalse
provider_name"Fireworks" 
latency294
moderation_latency(null)
generation_time209
finish_reason"stop" 
tokens_prompt60
tokens_completion9
native_tokens_prompt63
native_tokens_completion10
native_tokens_reasoning(null)
num_media_prompt(null)
num_media_completion(null)
origin"https://ncbench.com/" 
usage0.0000657

Evaluation details

Result Evaluator Details Meta Data
Fail Matches text
Case sensitive: 1993
n/a
Pass Matches Regex
/\b1993\b/
n/a
50.0000%