1. Affordable multimodal audio model

• Extremely low-cost audio + text model for production-scale usage.
• Ideal for startups and high-volume traffic apps.

2. Fast real-time performance

• Low latency suitable for responsive voice assistants, AI phone bots, IVR flows, and audio chat apps.
• Great when speed matters more than deep reasoning.

3. Audio input and audio output

• Accepts raw audio (speech, recordings, commands).
• Generates natural audio responses via the REST API.

4. Large 128K context window

• Handles long conversations, transcriptions, and extended instructions.
• Supports multi-step voice workflows or multi-part inputs.

5. Great for lightweight reasoning workloads

• Performs well for classification, instructions, Q&A, rewriting, and audio-driven tasks.
• Good for voice agents that don't need high-end reasoning like GPT-5.1.

6. Works across major endpoints

• Chat Completions, Responses API, Realtime API, Assistants, Batch.
• Supports streaming and function calling.

7. Scalable for commercial production

• Perfect for customer support hotlines, appointment bots, FAQ voice agents, or embedded voice UI in apps.
• Reliable and predictable output behavior given its price.

8. Preview model designed for experimentation

• Lets teams prototype voice-first features with minimal cost.
• Useful stepping-stone before upgrading to GPT-4o Audio or GPT-5 audio models.

1. Intelligence & Benchmark Performance

• Matches Claude 3 Opus (previous largest model) on many intelligence tasks.
• Surpasses Claude 3 Opus on multiple evaluations despite being a smaller, faster model.
• Major improvements across every skill category vs previous Haiku.

2. Coding Strength

• Scores 40.6% on SWE-bench Verified, outperforming:

  • Claude 3.5 Sonnet (original version)
  • GPT-4o
  • Many agent-driven systems

• Excellent for engineering assistants, agent coding tasks, and bug fixing.

3. Speed & Latency

• Same speed class as Claude 3 Haiku (ultra-fast).
• Ideal for real-time interactions, high request volumes, and UI responsiveness.

4. Tool Use & Instruction Following

• Better at following instructions than previous Haiku.
• Stronger at tool use accuracy, making it reliable for agents and workflows.

5. Best Use Cases

• High-volume, low-latency tasks
• User-facing products
• Sub-agent tasks in larger workflows
• Processing large structured datasets (pricing, inventory, purchase history)
• Rapid content or code generation where speed matters