Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 3 models | googleembeddinggemma-300mRemove × | intfloatIntfloat: Multilingual-E5-LargeRemove × | moondreamMoondream: Moondream 3 PreviewRemove × |
|---|---|---|---|
| What’s different | EmbeddingGemma is a 300M parameter multilingual open embedding model from Google DeepMind, designed for efficient deployment even on low-resource devices, producing high-quality text vector representations… | The multilingual-e5-large embedding model encodes sentences, paragraphs, and documents across over 90 languages into a 1024-dimensional dense vector space, delivering robust semantic embeddings optimized… | Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over… |
| Date | Sep 16, 2025Listed | Jun 4, 2024Listed | Oct 15, 2026Listed |
API prices USD · provider configuration shown below | |||
| Listed rate | See provider quotes | $0.01 in / $0.00 out/ 1M tokensOpenRouter · all routes ↗ | $0.30 in / $2.50 out/ 1M tokensPrimeIntellect · via OpenRouter · all routes ↗primeintellect/fp8 · quantization=fp8 |
| Provider endpoints | 0 endpoints → | 1 endpoint → | 2 endpoints → |
Specifications | |||
| Context window | 2K | 512 | 32K |
| Max output | — | 460 | 4.1K |
| Parameters | 300M | — | — |
| Image input | Not reported | Not reported | Supported |
| Reasoning | Not reported | Not reported | Supported |
| Open weights | Not reported | Supported | Supported |