customers
All customer stories
Top Voice AI companies are building with Assembly.
resources
Latest Release
Universal 3.5 Pro
Our new flagship speech-to-text model handles real-world audio the way it actually happens. Available for real-time and pre-recorded audio.
resources
From raw utterance to structured answer—fast open models for the cleanup pass, frontier models for the reasoning pass, one API key for both.
Speech-to-Text
Hi, Marissa. Um, quick recap on where we landed, uh, with the Q3 vendor review. So, uh, we got through 6 of the 8— oh no, 5 of the 8 assessments. Um, and the last 3, they’re waiting on security sign-off. Um, the big thing is that pricing came back higher than, than we modeled, something like, uh, between 12— no, 15% over. So I think we need to revisit The budget line before we commit to anything. Can you pull the original forecast for me? And I’ll put 30 minutes on the calendar.
LLM Gateway
Clean up dictated speech. Remove fillers (um, uh, like, you know), stutters, repetitions, false starts; on self-corrections keep only the final version. Fix punctuation, capitalization, sentence and paragraph breaks. Render spoken commands ("period", "new paragraph") as punctuation.
Final output
Hi Marissa, quick recap on where we landed with the Q3 vendor review. We got through five of the eight assessments. The last three are waiting on security sign-off. The big thing is that pricing came back higher than we modeled, something like 15% over. I think we need to revisit the budget line before we commit to anything. Can you pull the original forecast for me? I'll put 30 minutes on the calendar.
34 models from 5 providers, all OpenAI-compatible—switch by changing one string.
| Models | Provider | Input | Output |
|---|---|---|---|
| GPT-5 Nano | Open AI | $0.05 / 1M | $0.40 / 1M |
| GPT OSS 20B | Bedrock | $0.07 / 1M | $0.30 / 1M |
| Gemini 2.5 Flash Lite | Vertex | $0.10 / 1M | $0.40 / 1M |
| Qwen3.5 4B Fast | AssemblyAI | $0.10 / 1M | $0.50 / 1M |
| gemma-4-31b | Bedrock Mantle | $0.14 / 1M | $0.40 / 1M |
| GPT OSS 120B | Bedrock | $0.15 / 1M | $0.60 / 1M |
| Qwen3 32B | Bedrock | $0.15 / 1M | $0.60 / 1M |
| Qwen3 Next 80B A3B | Bedrock | $0.15 / 1M | $1.20 / 1M |
| Gemini 3.1 Flash Lite | Vertex | $0.25 / 1M | $1.50 / 1M |
| GPT-5 mini | Open AI | $0.25 / 1M | $2.00 / 1M |
| Gemini 2.5 Flash | Vertex | $0.30 / 1M | $2.50 / 1M |
| Gemini 3.5 Flash Lite | Vertex | $0.30 / 1M | $2.50 / 1M |
| Gemini 3.7 Flash | Vertex | $0.75 / 1M | $3.75 / 1M |
| Gemini 3.8 Flash | Vertex | $0.75 / 1M | $3.75 / 1M |
| GPT-5.6 Luna | Open AI | $1.00 / 1M | $6.00 / 1M |
| Haiku 4.5 | Bedrock | $1.00 / 1M | $5.00 / 1M |
| Gemini 2.5 Pro | Vertex | $1.25 / 1M | $10.00 / 1M |
| Gemini 3.5 Flash | Vertex | $1.25 / 1M | $9.00 / 1M |
| GPT-5 | Open AI | $1.25 / 1M | $10.00 / 1M |
| GPT-5.1 | Open AI | $1.25 / 1M | $10.00 / 1M |
| Gemini 3.6 Flash | Vertex | $1.50 / 1M | $7.50 / 1M |
| GPT-5.2 | Open AI | $1.75 / 1M | $14.00 / 1M |
| GPT-4.1 | Open AI | $2.00 / 1M | $8.00 / 1M |
| GPT-5.6 Terra | Open AI | $2.50 / 1M | $15.00 / 1M |
| Sonnet 4.5 | Bedrock | $3.00 / 1M | $15.00 / 1M |
| Sonnet 4.6 | Bedrock | $3.00 / 1M | $15.00 / 1M |
| Sonnet 5 | Bedrock | $3.00 / 1M | $15.00 / 1M |
| GPT-5.6 Sol | Open AI | $4.00 / 1M | $20.00 / 1M |
| GPT-5.5 | Open AI | $5.00 / 1M | $30.00 / 1M |
| Opus 4.5 | Bedrock | $5.00 / 1M | $25.00 / 1M |
| Opus 4.6 | Bedrock | $5.00 / 1M | $25.00 / 1M |
| Opus 4.7 | Bedrock | $5.00 / 1M | $25.00 / 1M |
| Opus 4.8 | Bedrock | $5.00 / 1M | $25.00 / 1M |
| Opus 5 | Bedrock | $5.00 / 1M | $25.00 / 1M |
Showing 34 of 34 models
No models match these filters.
Try clearing the search or widening the filters above.
Prices shown are for global routing. In-region (US/EU) pricing is 10% higher due to provider cost increases. Every model, with context windows and capabilities, is on the full catalog.
Ship faster, spend less on tokens, and stop losing users to provider outages.
Pay provider rates, not gateway rates. Competing gateways add 5% or more to every call.
Configure backup models per request. When a provider errors or stalls, your call still goes through.
Pay the exact same price as calling the model provider directly. No markup, no hidden fees, no minimum commitment. We make it simple.
Drop into any OpenAI SDK. Change a base URL and a model string, and everything keeps working.
Your LLM calls run where your transcription does. One less network hop on every turn.
Frontier models from OpenAI, Anthropic, and Google, plus open models we host ourselves. New ones added the day they launch.
Use our playground to upload a file, run a prompt over the transcript, and switch models to see how the answers differ.
