Sarvamsarvam familyActive

Sarvam-105B

Flagship Indian-language reasoning model for enterprise multilingual applications

JSON Embed chart Compare

Context

131,072

Max output

131,072

Input / 1M

$0.04

Output / 1M

$0.16

Providers

01 Profile 02 Capabilities 03 Pricing 04 History 06 Related

01Profile

Profile

Released

2026-02-18

Last updated

2026-03-06

Knowledge cutoff

—

Weights

Open

02Capabilities

Capabilities

Input

text

Output

text

Reasoning

Yes

Tool calling

Yes

Structured output

Attachments

Cache read

—

Cache write

—

03Pricing

Pricing across 2 providers

Per 1M tokens, USD. Sorted by input price. First-party row highlighted.

Provider	Provider model id	Input	Output	Cache read	Context
sarvamfirst-party	sarvam-105b	Not listed	Not listed	Not listed	131,072
fastrouter	sarvam/sarvam-105b	$0.04	$0.16	Not listed	131,072
nano-gpt	sarvam-105b	$0.05	$0.21	$0.03	131,072

04History

Change history

9+ recent events since 2026-02-18

2026-09-08Price upnano-gptinput / 1M$0.04→ $0.05

2026-09-08Price upnano-gptoutput / 1M$0.18→ $0.21

2026-08-03Contextnano-gptContext window131,072→ new

2026-06-10ContextfastrouterContext window128,000→ new

2026-06-10ContextfastrouterMax output128,000→ new

2026-06-09Addedfastrouteroffering→ new

2026-06-03Addednano-gptoffering→ new

2026-05-05ContextsarvamMax output2,048→ 131,072

2026-05-01Addedsarvamoffering→ new

Sarvam-105B

Profile

Capabilities

Pricing across 2 providers

Change history

Same family & alternatives