Last refreshed 2026-04-19. Next refresh: weekly.
Why use Arctic on Replicate API?
Replicate API offers Arctic with pay-as-you-go pricing at $0.65/1M input tokens. Replicate is a cloud-based platform that enables users to run machine learning models easily and efficiently.
Compare Arctic across 4 providers to find the best fit for your use caseSetup recipe
Python + curlpip install replicateexport REPLICATE_API_TOKEN=...import replicate
output = replicate.run(
"snowflake/snowflake-arctic-instruct",
input={"prompt": "Hello"}snowflake/snowflake-arctic-instructRequest example
import replicate
# reads REPLICATE_API_TOKEN from env
# snowflake/snowflake-arctic-instruct format: "owner/model-name" (latest version) or "owner/model-name:version-hash"
output = replicate.run(
"snowflake/snowflake-arctic-instruct",
input={"prompt": "Hello"}
)
# Output is a list or generator depending on the model
print("".join(output))Gotchas
- Use provider model ID "snowflake/snowflake-arctic-instruct", not the LLMReference slug "arctic".
- Replicate uses "owner/model-name" format (e.g. "meta/meta-llama-3-8b-instruct") for the latest version, or "owner/model-name:version-sha" to pin to a specific version. The REST endpoint splits owner and model-name into the path: /v1/models/{owner}/{model-name}/predictions.
- The examples expect REPLICATE_API_TOKEN; rename it only if your application config maps the new variable.
Compare Arctic Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| NVIDIA NIM | — | — |
| Microsoft Foundry | $2.00 | $2.00 |
| Together AI | $2.40 | $2.40 |
| Replicate API | $0.65 | $2.75 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.65 |
| Output tokens | $2.75 |
Capabilities
About Arctic
Snowflake Arctic is an advanced large language model tailored for enterprise applications by Snowflake AI Research. It features an innovative Dense-MoE Hybrid transformer architecture, combining a 10 billion parameter dense transformer with a 128 x 3.66 billion parameter MoE MLP, totaling 480 billion parameters but utilizing only 17 billion actively. This structure optimizes efficiency, particularly for tasks like SQL generation, coding, and instruction following. The model's training spanned a diverse dataset of 3.5 trillion tokens, focusing on enterprise needs. Despite its capabilities, Arctic's deployment presents challenges due to its size, and it remains vulnerable to inaccuracies with unclear inputs.