Price comparison · per model
Synap vs DeepInfra
The same open models, priced per model. Only models both of us list, and only where the numbers are checked today.
| Model | Synap | DeepInfra | Synap is |
|---|---|---|---|
| gpt-oss-20b | $0.017 / $0.085 | $0.03 / $0.14 | 43% / 39% lower |
| Gemma 4 26B | $0.04 / $0.21 | $0.07 / $0.34 | 43% / 38% lower |
| Qwen3 VL 30B | $0.14 / $0.55 | $0.15 / $0.60 | 7% / 8% lower |
| Muse Glimmer 30B | $0.28 / $1.10 | $0.30 / $1.20 | 7% / 8% lower |
USD per 1M tokens, input / output. DeepInfra prices checked 2026-10-31 22:40 UTC from OpenRouter, DeepInfra endpoint (Muse Glimmer 30B) · OpenRouter, DeepInfra endpoint (gpt-oss-20b) · deepinfra.com/pricing. Synap prices are read live from /v1/models, which is what the gateway bills. Prices only: context limits differ by model and are on each model page.
DeepInfra does not list Qwen3 Coder 30B or Qwen3 30B A3B Instruct 2507 on the pages checked. Our prices for those are on each model page.
Also on Synap
- Try any model in chat without an account
- Free tier: 100,000 tokens a day on qwen3-coder-30b with Google sign-in
- Pay by card, TON or USDC (x402); credits you buy never expire
- Zero data retention on every request
Switching is one line: set base_url to https://synap.linkrra.com/v1.