NVIDIA-tuned Llama 3.1 253B for maximum capability.
OffRail routes requests to the best providers that are able to handle your prompt size and parameters.
NVIDIA-tuned Llama 3.1 253B for maximum capability. You can access it through OffRail's OpenAI-compatible API with automatic provider routing, fallback, and cost analytics.
Pricing for Llama 3.1 Nemotron Ultra 253B on OffRail starts at $0.60 per million input tokens and $1.80 per million output tokens, depending on the provider. The pricing table above always reflects the current per-provider rates.
Llama 3.1 Nemotron Ultra 253B supports a context window of up to 128,000 tokens on its largest provider deployment.
Llama 3.1 Nemotron Ultra 253B is served by Nebius AI through OffRail. Requests are automatically routed to the best available provider, with fallback when a provider has issues.
Llama 3.1 Nemotron Ultra 253B supports soft JSON output, but not tool calling or strict JSON output schema.
Llama 3.1 Nemotron Ultra 253B was released on April 7, 2025.
Muse Spark 1.2
1,048,576 context$1.25 in / $4.25 out per 1M
Muse Spark 1.1
1,048,576 context$1.25 in / $4.25 out per 1M
Llama 4 Scout 17B Instruct
131,072 context$0.17 in / $0.59 out per 1M
Llama 4 Maverick 17B Instruct
1,048,576 context$0.24 in / $0.85 out per 1M
Llama 3.3 70B Instruct
131,072 context$0.13 in / $0.40 out per 1M
Llama 3.2 11B Instruct
128,000 context$0.07 in / $0.33 out per 1M