Pricingby GlyphNVIDIA NIM vs Helicone
Fees, free tiers and features, from each gateway's own pages.
- Markup on tokens
- Not published
- 0%
- Bring your own key
- No
- Yes, 0% fee
- Free tier
- Free inference endpoints for NVIDIA Developer Program members (prototyping only); 90-day NVIDIA AI Enterprise trial
- Hobby: 10,000 free requests, 1 GB storage
- Entry paid plan
- Usage based
- Pro, $79/mo
- Self-host
- No
- Yes
- Open source
- No
- Yes (Apache 2.0)
- Fallbacks
- Load balancing
- Caching
- Guardrails
- Observability
- Rate limits
- Budgets
- Prompts
- OpenAI-compatible
- Bring your own key
- Data residency
- Private networking
Monthly cost, 100M in and 20M out
DeepSeek V4.1 Flashn/aNot offered
gpt-oss-20bn/a$0.74
Llama Guard 4 12Bn/a$17.28
Gemma 3 12Bn/a$6.40
Claude Sonnet 5Not offeredNot offered