skip to content
gigarouter gigarouter
models / embeddings · coming soon

Arctic Embed M Long

Snowflake/snowflake-arctic-embed-m-long

published Apr 2024 · updated Dec 2024

Arctic Embed M Long is a text embedding model optimized for long-context retrieval, supporting up to 2048 tokens (or 8192 with rotary position embeddings).

est. price
~$0.008
/ 1M tokens · estimated, set at launch
API providers
0
downloads / mo
53.5K
license
apache-2.0

specs

TaskText Embedding / Retrieval
ArchitectureBased on nomic-embed-text-v1-unsupervised (BERT-style)
Parameters137 million
LicenseApache 2.0
Embedding Dimension768
Context LengthUp to 2048 tokens (8192 with RPE)

about this model

Snowflake/snowflake-arctic-embed-m-long is a text embedding model optimized for long-context retrieval tasks, supporting up to 2048 tokens without Rotary Position Embeddings (RPE) and scaling to 8192 tokens with RPE.

Model Architecture and Training

Based on the nomic-embed-text-v1-unsupervised architecture, this 137-million-parameter model produces 768-dimensional embeddings. It was trained using a multi-stage pipeline: first, pretraining on approximately 400 million query-document pairs with in-batch negative mining, followed by fine-tuning on roughly 1 million triplets of query, positive document, and hard negative documents derived from harmful mining. The training methodology is detailed in the Arctic-Embed technical report.

Retrieval Performance

On the MTEB Retrieval benchmark (NDCG@10), snowflake-arctic-embed-m-long achieves a score of 54.83, outperforming comparable models:

ModelMTEB Retrieval Score (NDCG@10)
snowflake-arctic-embed-m-long54.83
nomic-embed-text-v1.553.01
nomic-embed-text-v152.81

Key Strengths

  • Extended context window: supports 2048 tokens natively and up to 8192 tokens with RPE, making it suitable for long-document retrieval workloads.
  • Competitive accuracy: delivers retrieval quality near the larger snowflake-arctic-embed-l model (55.98) while using fewer parameters.
  • Open-source: released under the Apache-2.0 license with weights available for inspection and use.

Supported Formats

The model is available in ONNX, Safetensors, and Transformers.js formats, and is compatible with the Text Embeddings Inference framework.

best for

FAQ

What is Arctic Embed M Long best for?

It is best for retrieval tasks that require embedding long documents or passages, supporting up to 2048 tokens natively and up to 8192 tokens with rotary position embeddings.

How does Arctic Embed M Long compare to the base Arctic Embed M?

It has the same embedding dimension (768) but supports longer context (up to 2048/8192 tokens vs. 512 tokens) and has 137M parameters vs. 110M.

What is the license for Arctic Embed M Long?

The model is released under the Apache 2.0 license.

How do I call Arctic Embed M Long via the gigarouter API?

Use the OpenAI-compatible endpoint with your API key, sending texts to the embeddings endpoint.

What is the MTEB retrieval score of Arctic Embed M Long?

It achieves a score of 54.83 NDCG@10 on the MTEB Retrieval benchmark.

not yet live

We're benchmarking and onboarding Arctic Embed M Long as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.

related embeddings models

compare all →