Voyage 4 Nano
Voyage 4 Nano is Voyage's language model. A compact text embedding model from Voyage AI built on a bidirectional Qwen3 backbone (~340M parameters, 32k context), optimized for high-performance semantic search and retrieval.
Capabilities
Input1/5
✓
·
·
·
·
Output1/5
✓
·
·
·
·
Capabilities0/13
·
·
·
·
·
·
·
·
·
·
·
·
·
Other models
| Model | Tier | Released | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| DeepSeek R1 Distill Qwen3 8B | — | — | 131K | $0.200 | $0.200 |
| Qwen3 Embedding 0.6B | — | 33K | $0.010 | $0.000 | |
| Qwen3 Embedding 4B | — | 41K | $0.020 | $0.000 | |
| Qwen3 Embedding 8B | — | 41K | $0.020 | $0.000 | |
| Qwen3 14B | — | 131K | $0.060 | $0.200 | |
| Qwen3 32B | — | 131K | $0.050 | $0.100 | |
| Qwen3 8B | — | 131K | $0.035 | $0.138 | |
| Qwen3 4B Instruct | — | — | 262K | $0.010 | $0.030 |
| KwaiPilot KAT 32B Dev | — | — | 131K | $0.900 | $0.900 |
| Qwen3 0.6B | — | — | 41K | $0.100 | $0.100 |