Granite Embedding English R2
Aug 2025
Granite Embedding English R2Compare inon
Granite Embedding English R2 measured on 25 of 261 Embeddings benchmarks25
BenchmarkByScoreRank
No benchmark matches
self-host onlyprice
56.4MTEB English v2
Aug 2025released
768Dimensions
8K tokensMax input
enLanguages
Rankingconfiguration unstated?
Full ranking →···
44Bge Large En v1.558.245Granite Embedding English R258.246Bge M357.7Strong at
Alternatives?
ModelScoreCostPriceWhat changes
BetterQwen 3 Embedding 8BunstatedOpen+25%—$0.01Languages: not verified+1
OpenNemotron 3 Embed 8B BF16+51%——Memory (4-bit): 0.1 GB → 4.8 GBBetterNemotron 3 Embed 1B BF16Open+33%—$0.015Memory (4-bit): 0.1 GB → 0.7 GBVBetterVoyage 4 Lite+39%—$0.02Languages: not verified
BetterJina Embeddings v5 Text NanounstatedOpen+11%—$0.02Languages: not verifiedModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+25%—$0.01Languages: not verified+1Nemotron 3 Embed 1B BF16Open+33%—$0.015Memory (4-bit): 0.1 GB → 0.7 GBVVoyage 4 Lite+39%—$0.02Languages: not verifiedJina Embeddings v5 Text NanounstatedOpen+11%—$0.02Languages: not verified
Text Embedding 3 Smallunstated+5%—$0.02Languages: not verifiedModelScoreCostPriceWhat changes
Nemotron 3 Embed 8B BF16Open+51%——Memory (4-bit): 0.1 GB → 4.8 GBOOcten Embedding 8BOpen+43%——Memory (4-bit): 0.1 GB → 4.5 GBOOcten Embedding 8B INT8Open+42%——Memory (4-bit): 0.1 GB → 4.5 GBOOcten Embedding 4BOpen+38%——Memory (4-bit): 0.1 GB → 2.4 GBOOcten Embedding 4B INT8Open+37%——Memory (4-bit): 0.1 GB → 2.4 GBModelScoreCostPriceWhat changes
Granite Embedding 311M Multilingual R2unstatedOpen-9%——Memory (4-bit): 0.1 GB → 0.2 GBGranite Embedding Small English R2unstatedOpen-13%——Granite Embedding 97M Multilingual R2unstatedOpen-17%——Granite Embedding 107M MultilingualunstatedOpenn/a——Max input: 8K tokens → 1K tokensGranite Embedding 125M EnglishunstatedOpenn/a——Max input: 8K tokens → 1K tokensGranite Embedding 278M MultilingualunstatedOpenn/a——Max input: 8K tokens → 1K tokens+1Granite Embedding 30M EnglishunstatedOpenn/a——Max input: 8K tokens → 1K tokensAll IBM models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
in Embeddings & RetrievalBEIR · Retrieval · English53.1#66
FAQ
How much does Granite Embedding English R2 cost?›
There is no per-unit API price: self-host only.
Is Granite Embedding English R2 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Granite Embedding English R2, and what is it best at?›
#47 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.
Is Granite Embedding English R2 better than Bge Large En v1.5?›
On RTEB Finnish beta · Mean over tasks · English, Granite Embedding English R2 scores 58.2 vs 58.2 — about the same.
When was Granite Embedding English R2 released? Is there a newer version?›
Released 2025-08-15. It is the current version in its line.
Is Granite Embedding English R2 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Granite Embedding English R2 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Granite Embedding English R2?›
Granite Embedding English R2 is an open-weights embeddings & retrieval model from IBM, released Aug 2025.
What are the best alternatives to Granite Embedding English R2?›
Qwen 3 Embedding 8B (better), Nemotron 3 Embed 8B BF16 (open weights).