Lowest listed starting rate
OpenAI text-embedding-3-small$0.02 input / $0 output per 1M tokens
Per 1M input tokens; batch pricing is supported.
Pricing comparison
text-embedding-3-small vs Voyage 4 Pricing Calculator
Compare estimated monthly cost, unit pricing, context windows, rate limits, and developer capabilities for text-embedding-3-small and Voyage 4.
Cheapest eligibleCalculating
Balanced optionCalculating
CoverageCalculating
| Provider / model | Estimated monthly cost | Unit price | Context | Rate limit | Signal |
|---|
Excluded or custom-priced models
| Provider / model | Status | Reason |
|---|
Side-by-Side Pricing
| Provider | Model | Category | Current pricing | Context | Capabilities |
|---|---|---|---|---|---|
| OpenAI | text-embedding-3-small Per 1M input tokens; batch pricing is supported. |
embedding | $0.02 input / $0 output per 1M tokens | 8,191 | batch api |
| Voyage AI | Voyage 4 First 200M tokens are free; then $0.06 per 1M. Batch is 33% off. |
embedding | $0.06 input / $0 output per 1M tokens | 32,000 | batch api |
Recommended Starting Points
Balanced comparison pick
Voyage AI Voyage 4$0.06 input / $0 output per 1M tokens
First 200M tokens are free; then $0.06 per 1M. Batch is 33% off.
Pricing Details
| Provider | Model | Category | Current pricing | Context | Capabilities |
|---|---|---|---|---|---|
| OpenAI | text-embedding-3-small Per 1M input tokens; batch pricing is supported. |
embedding | $0.02 input / $0 output per 1M tokens | 8,191 | batch api |
| Voyage AI | Voyage 4 First 200M tokens are free; then $0.06 per 1M. Batch is 33% off. |
embedding | $0.06 input / $0 output per 1M tokens | 32,000 | batch api |
Limits and Notes
Rates checked 2026-11-10. Tiered, batch, cached-input, peak/off-peak, and regional pricing can differ from the standard planning rate shown here. Confirm the official source before committing budget.
FAQ
How is monthly cost calculated?
Token APIs use input and output token prices. Per-request APIs use monthly requests. Character-priced TTS uses monthly characters, while token-priced speech models use text and audio token volumes. Vector database estimates use storage and read/write units when the provider publishes them.
Why are some models excluded?
The calculator excludes models that miss required context, rate-limit, streaming, JSON, or tool-calling requirements. Custom or purely resource-based pricing is also excluded from numeric ranking.
Are prices guaranteed?
No. This is a planning dataset with official source links. Model availability, promotions, and account-tier limits can change.
Data Sources and Last Checked
- OpenAI official pricing · checked 2026-11-10
- Voyage AI official pricing · checked 2026-11-10