Speech 2.8 Turbo
speech-2.8-turboPricing
List prices at Replicate, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $60.00 |
What it costs a month
At Replicate list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out$18,000
- RAG over documents2,000 requests a day, 8K in / 0.6K out$28,800
- Coding agent1,000 requests a day, 40K in / 2K out$72,000
- Overnight batch20,000 requests a day, 4K in / 0.4K out$144,000
Where to run it
1 host serve Speech 2.8 Turbo. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
ReplicateInference speech-2.8-turbo | $60.00 / - | speech-2.8-turbo |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- Text
- Produces
- Audio
- Open weights
- License: Apache-2.0Runtimes: SGLang, vLLM
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than Replicate.
Smaller Independent models
No cheaper Independent model keeps its key capabilities.
Other frontier labs
- 90% less
- 95% less
- 96% less
- 97% less
Open-weight picks
- 99% less
Call it
Using Replicate's API.
from litellm import completion
response = completion(
model="replicate/speech-2.8-turbo",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Speech 2.8 Turbo cost?
On Replicate, Speech 2.8 Turbo costs $60.00 per million input tokens.
Which APIs offer Speech 2.8 Turbo?
Speech 2.8 Turbo runs on 1 API: Replicate.
Is Speech 2.8 Turbo being retired?
No retirement date is scheduled for it on any API we track.
What are cheaper alternatives to Speech 2.8 Turbo?
Models that cost less and keep most of what it can do: Gemini 2.5 Pro TTS on Vertex AI (90% less), GPT-4o Mini TTS on OpenAI (95% less) and Qwen Audio 3.1 TTS Flash on Alibaba DashScope (99% less).
What is Speech 2.8 Turbo's API model id?
Use speech-2.8-turbo with Replicate.
Sources
Raw catalog entry for speech-2.8-turbo on Replicate
{
"input_cost_per_token": 0.00006,
"litellm_provider": "replicate",
"mode": "audio_speech",
"source": "https://replicate.com/minimax/speech-2.8-turbo"
}