LIVE Cheapest paid: Granite 4.0 Micro $0.017/Mtok in 174 models tracked Updated Jul 26, 2026
Jul 26, 2026
ModelPriceWatch$/Mtok
Pricing / Compare / Granite Embedding 278M Multilingual vs Embed 4

Granite Embedding 278M Multilingual vs Embed 4

Side-by-side comparison of API pricing, specs, benchmarks, and capabilities

Granite Embedding 278M Multilingual is 11.7% cheaper on blended cost ($0.053 vs $0.060/Mtok)
 
Iby IBM
Cby Cohere
Overview
StatusCurrent embedding Open weights Current embedding
Released Jun 1, 2025 Jun 1, 2025
Pricing per million tokens
Input $0.106/Mtok $0.120/Mtok
Output $—/Mtok $—/Mtok
Blended avg $0.053/Mtok $0.060/Mtok
Specifications
Context window tokens tokens
Parameters 278M Proprietary
Speed (TPS)
Modalities
Input
text
textimage
Providers
Available from
IBM — $0.106/$—/Mtok
Cohere — $0.120/$—/Mtok

Cost at scale — 1M tokens (50/50 input/output)

VolumeGranite Embedding 278M MultilingualEmbed 4Savings
1M tokens $0.05 $0.06 $0.01 (16.7%)
10M tokens $0.53 $0.6 $0.07 (11.7%)
100M tokens $5.3 $6 $0.7 (11.7%)
1000M tokens $53 $60 $7 (11.7%)

Summary

Granite Embedding 278M Multilingual by IBM costs $0.106/Mtok input and $—/Mtok output, with a —-token context window. It supports text input.

Embed 4 by Cohere costs $0.120/Mtok input and $—/Mtok output, with a —-token context window. It supports text, image input.

On a blended cost basis, Granite Embedding 278M Multilingual is 11.7% cheaper than Embed 4.

Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.