Sarvam-30B

Sarvam-30B is an efficient Indian-language reasoning model for chat, coding, and multilingual work, as described by Sarvam AI. It is part of the sarvam line, which also includes the larger Sarvam-105B released on the same date. It has a context window of 65536 tokens and a maximum output of 65536 tokens. It accepts and returns text. It supports reasoning mode and tool calling. The weights are open. It is served by 2 hosts. Pricing starts at $0.02 per million input tokens and $0.10 per million output tokens. The knowledge cutoff is not stated.

First version of its line in the registry.

• ·Reasoning mode and tool calling supported
• ·Context window of 65536 tokens
• ·Max output of 65536 tokens
• ·Open weights, served by 2 hosts
• ·Input price from $0.02 per million tokens

• Reach for it for Indian-language reasoning tasks
• Reach for it for coding assistance in multilingual contexts
• Reach for it for chat applications needing tool calling
• Reach for it for long-context generation up to 65536 tokens

2 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window and output limit?

The context window is 65536 tokens, and the maximum output is also 65536 tokens.

Is the model open weights?

Yes, the weights are open. The model is served by 2 hosts.

What is the pricing?

Pricing starts at $0.02 per million input tokens and $0.10 per million output tokens.