Harrier Models by Microsoft Research
3 models2026Up to 33k ctx
About
Microsoft Bing's family of open-source multilingual text embedding models. Decoder-only architecture with last-token pooling. Three sizes: 270M, 0.6B, 27B. All support 32K context and 94+ languages.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
3 in view
Harrier OSS v1 27BCurrent
Use when the workload needs embedding, 33k context, and 27B parameters.
2026-03embedding33k context27B parameters
Harrier OSS v1 0.6BCurrent
Use when the workload needs embedding, 33k context, and 600M parameters.
2026-03embedding33k context600M parameters
Harrier OSS v1 270MCurrent
Use when the workload needs embedding, 33k context, and 270M parameters.
2026-03embedding33k context270M parameters
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Harrier OSS v1 27B | Use when the workload needs embedding, 33k context, and 27B parameters. | 2026-03 | embedding33k context27B parameters | Current |
| Harrier OSS v1 0.6B | Use when the workload needs embedding, 33k context, and 600M parameters. | 2026-03 | embedding33k context600M parameters | Current |
| Harrier OSS v1 270M | Use when the workload needs embedding, 33k context, and 270M parameters. | 2026-03 | embedding33k context270M parameters | Current |
Release Timeline
1 release group2026-03
3 current
Harrier OSS v1 0.6B
Currentembedding33k context600M parameters
Harrier OSS v1 270M
Currentembedding33k context270M parameters
Harrier OSS v1 27B
Currentembedding33k context27B parameters
Specifications(3 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| Harrier OSS v1 27B | 2026-03 | 33k | 27B |
| Harrier OSS v1 0.6B | 2026-03 | 33k | 600M |
| Harrier OSS v1 270M | 2026-03 | 33k | 270M |
Frequently Asked Questions
- What is Harrier used for?
- Harrier is used for embedding and coding. The family description and listed model capabilities point to those workloads as the best fit.
- How does Harrier compare to Azure Speech Services?
- Harrier by Microsoft Research is strongest where you need embedding, while Azure Speech Services by Microsoft Research is the closest related family to check for audio. Harrier has 3 listed variants and reaches up to 33k context, so compare the specs and pricing tables before choosing a production model.
- Which Harrier model should I use?
- If price is the main constraint, use the pricing table first because Harrier does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate Harrier OSS v1 27B with 33k context.
