LLM comparison workbench
Compare AI models side by side
Run one prompt across selected AI models, compare their responses, and inspect latency, token usage, and throughput in a single benchmark.
Using fallback API keys
You're currently using shared API keys with rate limits. For better performance, you canadd your own API keys.
How BotBattle works
1. Select and run
Select AI models and run the same prompt across each one.
2. Inspect the results
Inspect side-by-side responses with latency, token usage, and throughput measurements.
3. Analyze the comparison
With two or more results, run comparative analysis across the responses.
Some models use shared free-tier access; others require your own provider keys. Keys are stored in browser memory or localStorage according to your settings mode. When you test a key or run a benchmark, the selected key is sent through BotBattle's API to the provider.