Sign in
Access the GenAI Chat application
Sign in
GenAI Chat
Chat
Compare
Evaluate
PDF Q&A
Claude Sonnet 4.6
Claude Opus 4.6
Llama 4 Maverick
Amazon Nova Pro
GLM 5
DeepSeek V3.2
★ Llama Doctor 3B (CMI)
Send
Powered by Amazon Bedrock · Shift+Enter for newline
Model A
Claude Sonnet 4.6
Claude Opus 4.6
Llama 4 Maverick
Amazon Nova Pro
GLM 5
DeepSeek V3.2
★ Llama Doctor 3B (CMI)
Model B
Claude Sonnet 4.6
Claude Opus 4.6
Llama 4 Maverick
Amazon Nova Pro
GLM 5
DeepSeek V3.2
★ Llama Doctor 3B (CMI)
Compare
Both models receive the same prompt simultaneously
Document
📄
Drop PDF here or click to upload
Mode
Single model Q&A
Cache comparison (write → read)
Cross-model cache comparison
Prompt A/B test
Model
Claude Sonnet 4.6
Amazon Nova Pro
Claude Opus 4.6
Llama 4 Maverick
Enable prompt caching
Prompt Template
Structured — document analyst
Direct — concise answer
Chain-of-Thought — step-by-step
Socratic — guided discovery
Model
Claude Sonnet 4.6
Amazon Nova Pro
Template A
Structured
Direct
Chain-of-Thought
Socratic
Template B
Direct
Structured
Chain-of-Thought
Socratic
Session Stats
Queries
0
Total tokens
0
Total cost
$0.00000
Cache writes
0
Cache hits
0
Hit rate
—
Cost saved
$0.00000
Avg cached latency
—
Avg uncached latency
—
📄
Upload a PDF, then ask questions about it.
Text extracted client-side · Prompt caching via Bedrock native API · Metrics from actual API responses
Ask
Test Cases
12 cases
▶ Run All Test Cases
Select a test case and click Run,
or click "Run All" to evaluate all cases across all 4 models.