StatusOperational Live status
LLMAPI discountUp to −25%
Models on LLMAPI5
From$0.15 / 1M blended
RegionsUS
Best uptime (24h)99.80%
Verdict
Replicate hosts a large library of community-published models — language, image, video, and audio — all callable through one API. Through LLMAPI you reach 5 Replicate models on the same OpenAI-compatible endpoint as every other provider, so switching between them is a one-string change.
Good fit for
- Text Generation
- Image Generation
- Vision / Image Understanding
- Speech-to-Text
- Multimodal (any-to-any)
- Best for High Volume Production
Consider another provider for
- No code generation models in this catalog
- No video generation models in this catalog
- No video understanding models in this catalog
- No text-to-speech models in this catalog
Key facts
Where a provider does not document something we say so rather than guess. Everything LLMAPI records about Replicate, in one table.
Overview | |
| Provider | Replicate |
| Website | replicate.com |
| Best for | Prototyping with thousands of community models |
| Models on LLMAPI | 5 |
| Popular models | Llama 3.3 70B, FLUX.1, Whisper Large v3, SDXL, Thousands of community models via one API |
Access & routing | |
| API compatibility | OpenAI-compatible — /v1/chat/completions |
| Model ID prefix | replicate/<model> |
| Regions | US |
| Fallback | Configurable — on 429 or 5xx the router switches to any other model or provider on your list |
| Billing | One invoice across every provider, with per-key spend limits |
| Volume discount | Eligible |
Performance (median across listed models) | |
| Blended price / 1M | $0.15 |
| Uptime (24h) | 99.80% (median across listed models) |
Modality | |
| Tags | |
Price & performance | |
| Tags | |
Deployment | |
| Tags | |
Trust & safety | |
| Tags | |
Privacy & data | |
| Inference location | US |
| Data residency | Not documented by the provider |
| Default retention | Prompts and outputs deleted 1 hour after an API prediction finishes |
| Zero-data retention | Not documented by the provider |
| Training on API data | Not documented by the provider |
| Human review | Not documented by the provider |
Compliance & lifecycle | |
| Certifications | Not documented by the provider |
| Deprecation notice | Not documented by the provider |
| Operational status | Operational Live status |
Prices and context from the OpenRouter public catalogue; uptime and output speed from OpenRouter provider endpoints; quality scores from LiveBench. Refreshed nightly — last updated 2026-09-29. — means no figure is published for that model.
Best Replicate models
A quick view of how Replicate 's 5 models compare on intelligence, output speed and price — so you can pick the right one for your use case.
Most intelligent
#1Thousands of community models via one API43
#2SDXL42
#3Whisper Large v335
#4Llama 3.3 70B26
Intelligence index · 5 models
Fastest
#1SDXL1,024 t/s
#2Whisper Large v3853 t/s
#3FLUX.1748 t/s
#4Thousands of community models via one API722 t/s
Output tokens / second · 5 models
Lowest price
#1Whisper Large v3$2.80
#2SDXL$3.94
#3FLUX.1$8.16
#4Llama 3.3 70B$10.20
Blended price per 1M tokens · 5 models
Compared with other hosts
How Replicate sits against the rest of the catalog, using the same measurements shown on every provider page.
- Price. Median blended price of $8.16 per 1M tokens — 33% above the $6.13 median across all 50 providers in the catalog.
- Throughput. Median output speed of 748 tokens/s, 115% faster than the 348 tok/s catalog median.
- Catalog depth. 5 models listed, against a catalog average of 5 per provider.
- Integration. Identical to every other provider here — same endpoint, same SDK, same keys, so a switch costs one string change.
Median blended price per 1M tokens
lower is better| Replicate | $8.16 | |
| Agnes AI | $8.20 | |
| Celeris | $8.21 | |
| Multiverse Computing | $8.44 | |
| Together AI | $7.80 | |
| OpenAI | $7.75 |
Median output speed
higher is better| Replicate | 748 tok/s | |
| Agnes AI | 274 tok/s | |
| Celeris | 661 tok/s | |
| Multiverse Computing | 204 tok/s | |
| Together AI | 781 tok/s | |
| OpenAI | 348 tok/s |
Figures are medians across the models listed on each provider page and are refreshed with the catalog.
Models & pricing
Every Replicate model reachable with your LLMAPI key. Click a column heading to sort.
| Model | API model name | Uptime 24h | Price / 1M | Context |
|---|---|---|---|---|
| Llama 3.3 70B | replicate/llama-3-3-70b | 99.80% | $ 0.15 | 131 k |
| FLUX.1 | replicate/flux-1 | — | — | — |
| Whisper Large v3 | replicate/whisper-large-v3 | — | — | — |
| SDXL | replicate/sdxl | — | — | — |
| Thousands of community models via one API | replicate/thousands-of-community-models-via-one-api | — | — | — |
Price is a blended per-1M-token figure; latency is time to first token and throughput is output tokens per second, measured on the LLMAPI edge.
Cost calculator
Pick a model, enter your traffic, and see the monthly bill at LLMAPI rates.
Estimates use the blended per-1M-token price shown in the table above. Word counts differ by language — Cyrillic and CJK text uses 2–3× more tokens per word.
Replicate via LLMAPI
Same models, same features, one key. Prefix the model ID with replicate/ and point your OpenAI client at our base URL.
# Python · OpenAI SDK from openai import OpenAI client = OpenAI( base_url="https://api.llmapi.ai/v1", api_key="LLMAPI_KEY", ) r = client.chat.completions.create( model="replicate/llama-3-3-70b", messages=[{"role": "user", "content": "Hello"}], )
| Pricing | Eligible for LLMAPI volume discounts; billed on one invoice with every other provider. |
| Features | Full pass-through — tool calling, JSON schema, streaming, vision and reasoning behave exactly as on Replicate’s own API. |
| Fallback | Configurable. On 429 or 5xx the router switches to any model or provider on your fallback list. |
| Regions | US |
| Your data | LLMAPI stores request metadata only — model, token counts, latency and status. No prompt or completion content. |
Capabilities & use cases
Each tag below is its own catalog page listing every provider that shares it.
Developer resources
Where to go next.
Join thousands of developers building on one API key
Every provider in the catalog, one key, one invoice.
Questions
How do I call Replicate models through LLMAPI?
Point any OpenAI-compatible client at https://api.llmapi.ai/v1 and use the model ID replicate/llama-3-3-70b. No other change is needed.
Does Replicate cost more through LLMAPI?
No. This provider is eligible for LLMAPI volume discounts, so heavy usage lands below list price.
Which regions are used?
US
What happens if the provider returns an error?
On 429 or 5xx responses the router falls back to any other model or provider on your configured list, so requests keep succeeding.
How many Replicate models are available?
5 at the moment, all listed in the models table above and updated as the provider ships new ones.
Related providers
Providers covering similar use cases.