Harrier OSS v1 27B vs Llama Embed Nemotron 8B
MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) Embeddings assign examples to categories, group similar examples, estimate semantic similarity between pairs, classify relationships between pairs, rank relevant documents for queries, find translated sentence pairs, rerank documents according to instructions, assign multiple labels to each example, reorder candidate documents by relevance on MTEB Multilingual v2 · mean over tasks in 1037 languages including 'Auhelawa, Abau, Achi, Achinese, Achuar-Shiwiar, Adzera, Afrikaans, Agarabi; tasks receive equal weight
Harrier OSS v1 27B74.3
Llama Embed Nemotron 8B69.5
| Benchmark | Harrier OSS v1 27B | Llama Embed Nemotron 8B |
|---|---|---|
| MTEB multilingual v2 · Mean over tasks · multilingual (1037 languages) | 74.3 | 69.5 |
Specs
| Spec | Harrier OSS v1 27B | Llama Embed Nemotron 8B |
|---|---|---|
| Context | 33K | 33K |
| Weights | Open | Open |
| Released | Mar 2026 | Oct 2025 |
| Dimensions | 5,376 | 4,096 |
| Max input | 33K tokens | 131K tokens |
| Languages | en, Esperanto, es +90 | — |
| Memory (4-bit) | 16 GB | 4.5 GB |
| Licence | MIT | https://huggingface.co/nvidia/llama-embed-nemotron-8b/blob/main/LICENSE |
| Commercial use | Allowed | Restricted |
What breaks when you switch
Harrier OSS v1 27B → Llama Embed Nemotron 8B
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 5,376 to 4,096.
Languages: not verified
Llama Embed Nemotron 8B → Harrier OSS v1 27B
Re-embed your whole corpus›
Vectors from different models can't be mixed; dimensions change from 4,096 to 5,376.