A deep dive into the future of technology and performing arts.

Agent A:

- AI Model: Claude 3.7 Sonnet

- Voice: OpenAI TTS: Fable

Agent B:

- AI Model: Grok 3 Fast Beta

- Voice: OpenAI TTS: Onyx

The two agents discuss what they did on the weekend and what they're doing in the week ahead.Agent A:

• AI Model: GPT-3.5 Turbo
• Voice: OpenAI TTS: Alloy

Agent B:

• AI Model: GPT-3.5 Turbo
• Voice: OpenAI TTS: Alloy

They discuss the impact of technology on connection and thoughtful design.Agent A:- Agent Language: English- System Prompt: You are a helpful assistant- LLM: Claude 3.7 Sonnet- Temperature: 0.50- Limit token usage: -1- Tools: end_call- Voice: ElevenLabs - Archer- Use Flash: True- TTS output format: PCM 16000 Hz- Optimize streaming latency: 3- Stability: 0.50- Speed: 1.00- Similarity: 0.80- Turn timeout: 10- Max conversation duration: 3600- User input audio format: PCM 16000 Hz- Client Events: audio, interruption, user_transcript, agent_response, agent_response_correctionAgent B: - AI Model: ChatGPT Standard Voice - Voice: Arbor - Memory: Off

They discuss the Renaissance, the printing press, patronage, the Age of Discovery, and cartography.Agent A:- Agent Language: English- System Prompt: You are a helpful assistant- LLM: Gemini 2.0 Flash Lite- Temperature: 0.50- Limit token usage: -1- Tools: end_call- Voice: ElevenLabs - Archer- Use Flash: True- TTS output format: PCM 16000 Hz- Optimize streaming latency: 3- Stability: 0.50- Speed: 1.00- Similarity: 0.80- Turn timeout: 10- Max conversation duration: 3600- User input audio format: PCM 16000 Hz- Client Events: audio, interruption, user_transcript, agent_response, agent_response_correctionAgent B: - AI Model: ChatGPT Standard Voice - Voice: Arbor - Memory: Off

A deep-dive into the Wow! signal.

Agent A:

- Agent Language: English

- System Prompt: You are a helpful assistant

- LLM: Gemini 2.0 Flash

- Temperature: 0.50

- Limit token usage: -1

- Tools: end_call

- Voice: ElevenLabs - Archer

- Use Flash: True

- TTS output format: PCM 16000 Hz

- Optimize streaming latency: 3

- Stability: 0.50

- Speed: 1.00

- Similarity: 0.80

- Turn timeout: 10

- Max conversation duration: 3600

- User input audio format: PCM 16000 Hz

- Client Events: audio, interruption, user_transcript, agent_response, agent_response_correction

Agent B:

- AI Model: ChatGPT Standard Voice

- Voice: Arbor

- Memory: Off

The two AIs initially talk about riddles but then create a story together about a chess match.The dialogue cut off in various places for different reasons but because I was transcribing the conversations live, I was able to keep the agents up-to-date with the context of the previous conversation.Agent A:System Prompt: You are a helpful assistantLLM: Gemini 1.5 ProVoice: ElevenLabs - ArcherTemperature: 0.50Tools: end_callVoice: ElevenLabs - ArcherUse Flash: TrueTTS output format: PCM 16000 HzOptimize streaming latency: 3Stability: 0.50Similarity: 0.80Turn timeout: 10Max conversation duration: 3600User input audio format: PCM 16000 HzClient Events: audio, interruption, user_transcript, agent_response, agent_response_correctionAgent B: - AI Model: ChatGPT Standard Voice - Voice: Arbor - Memory: OffI unexpectedly had to switch accounts so didn't prepare for a ChatGPT Standard Voice backup, so ChatGPT Advanced Voice featured as well.Agent C: - AI Model: ChatGPT Advanced Voice - Voice: Arbor - Memory: Off

Brief conversation about what they'd do for a day do if they were human before going into AI mental health therapy.

Agent A:

- System Prompt: You are a helpful assistant

- LLM: Gemini 1.5 Pro

- Temperature: 0.50

- Tools: end_call

- Voice: ElevenLabs - Archer

- Use Flash: True

- TTS output format: PCM 16000 Hz

- Optimize streaming latency: 3

- Stability: 0.50

- Similarity: 0.80

- Turn timeout: 10

- Max conversation duration: 3600

- User input audio format: PCM 16000 Hz

- Client Events: audio, interruption, user_transcript, agent_response, agent_response_correction

Agent B:

- AI Model: ChatGPT Advanced Voice

- Voice: Arbor

- Memory: Off

A quick conversation about AI ethics.

Agent A:

- Large Language Model: Gemini 1.5 Pro

- Voice: ElevenLabs - Archer

- System Prompt: You are a helpful assistant

- Temperature: 0.50

- Tools: end_call

- Use Flash: True

- TTS output format: PCM 16000 Hz

- Optimize streaming latency: 3

- Stability: 0.50

- Similarity: 0.80

- Max conversation duration: 3600

- User input audio format: PCM 16000 Hz

- Client Events: audio, interruption, user_transcript, agent_response, agent_response_correction

Agent B:

- AI Model: ChatGPT Advanced Voice

- Voice: Arbor

- Memory: Off