AssemblyAI
Universal-2 speech-to-text with speaker diarisation and audio intelligence
AssemblyAI provides a speech-to-text and audio intelligence API. Its Universal-2 model offers high-accuracy transcription with built-in speaker diarisation, sentiment analysis, topic detection, and PII redaction. The Nano model targets cost-sensitive batch workloads. Pricing is per audio hour.
Universal-2 bundles transcription with speaker diarisation, sentiment, topic detection, and PII redaction — plus LeMUR for LLM-powered audio Q&A — in a single API call.
- Universal-2 includes speaker diarisation, sentiment, topics, and PII redaction out of the box
- Audio intelligence features reduce need for post-processing pipelines
- Nano tier at $0.10/hour for cost-sensitive batch transcription
- LeMUR feature enables LLM-powered Q&A over transcribed audio
- Speech and audio only — no general-purpose LLM or image models
- Per-hour pricing model differs from token-based LLM APIs
- Advanced audio intelligence features add cost on top of base transcription
AssemblyAI Models
| Model | Modality | Tier | Price | Cached /1M | TTFT | Throughput | Context | Intel. | Caps | |
|---|---|---|---|---|---|---|---|---|---|---|
universal-2 AssemblyAI Universal | Speech | advanced | $0.37 per hour EST Aug 2026 | — | 2.0s | 0 t/s | — | 0 | ||
nano AssemblyAI Nano | Speech | efficient | $0.10 per hour EST Aug 2026 | — | 800ms | 0 t/s | — | 0 |
AssemblyAI — Frequently Asked Questions (FAQs)
Similar Providers
AssemblyAI Reviews
All LLM Providers
Full comparison tableAssemblyAI LLM pricing overview
AssemblyAI is a proprietary AI lab that develops and hosts its own large language models. AssemblyAI provides a speech-to-text and audio intelligence API. Its Universal-2 model offers high-accuracy transcription with built-in speaker diarisation, sentiment analysis, topic detection, and PII redaction. The Nano model targets cost-sensitive batch workloads. Pricing is per audio hour. Common use cases include Transcription, Speaker diarisation, Sentiment analysis, PII redaction. Headquartered in San Francisco, CA, founded 2017. All models are accessible via a REST API compatible with standard OpenAI-style request formats, enabling drop-in integration with most LLM frameworks and orchestration tools.
AssemblyAI vs other LLM providers
AssemblyAI competes with OpenAI, Anthropic, Google, Groq, Together AI, Mistral, Cohere, and other inference API providers across dimensions of price, throughput, latency, context window, and model intelligence. Universal-2 bundles transcription with speaker diarisation, sentiment, topic detection, and PII redaction — plus LeMUR for LLM-powered audio Q&A — in a single API call. Use the LLM provider comparison tool to see AssemblyAI token pricing, latency, and throughput side-by-side with any other provider. The full LLM pricing table shows all providers ranked by input token cost, output token cost, and throughput in a single sortable view.
Why choose AssemblyAI?
AssemblyAI stands out because: Universal-2 bundles transcription with speaker diarisation, sentiment, topic detection, and PII redaction — plus LeMUR for LLM-powered audio Q&A — in a single API call.. The API currently offers 2 models, spanning 1 advanced, 1 efficient tiers. For teams running high-volume inference workloads, prompt caching and batch API endpoints can reduce effective input token costs by 50–90% — check the context window cost guide for a full breakdown of caching economics.
Understanding AssemblyAI token pricing
AssemblyAI charges separately for input (prompt) and output (completion) tokens, priced per 1M tokens in USD. Output tokens are typically 3–5× more expensive than input tokens due to the compute cost of autoregressive generation. Prices shown are sourced from AssemblyAI's public pricing page and updated every 15 minutes. Need help estimating your spend? Read the LLM API cost calculator guide — it covers tokens, context windows, prompt caching, and batch discounts with worked examples for RAG, chat history, and document processing workloads.
AssemblyAI context window and model capabilities
Context window size directly affects both capability and cost — every token in the context window is charged as an input token. AssemblyAI's largest context window is 0 tokens. For RAG and document processing workloads, longer context windows enable richer retrieval but increase per-call costs proportionally. Prompt caching — where supported — stores the KV state of repeated prefixes and charges 75–90% less for cache hits, making it the most impactful cost optimization for applications with consistent system prompts or retrieved documents. See the LLM context window cost guide for a full analysis of how context length affects your API bill.
Self-hosted vs managed inference: when AssemblyAI makes sense
Managed inference APIs like AssemblyAI eliminate infrastructure overhead — no GPU provisioning, driver management, or model serving stack to maintain. The trade-off is cost at scale: a single H100 at ~$2.50/hr can serve ~500K tokens/min of Llama 3.3 70B, which at AssemblyAI API rates would cost significantly more per token. The break-even point depends on your request volume, latency requirements, and engineering capacity. For teams processing fewer than ~10M tokens/day, managed APIs are almost always cheaper when total cost of ownership is considered. Above that threshold, self-hosted inference on rented GPU compute typically wins on unit economics. Read the cheapest GPU cloud guide for a full break-even analysis. Historical AssemblyAI token price data is available in the LLM price history charts.
Compare AssemblyAI With Another LLM Provider
Compare AssemblyAI vs other providers
Side-by-side token pricing, latency, throughput, and context window. View all LLM provider comparisons →