Developer Pricing
Build with Riven. Simple monthly plans, predictable costs.
Every plan includes a monthly token allocation shared across input and output — no per-call billing, no surprise invoices. Pick a plan, get your API key, and start building.
Looking for Chat, Relay, or team plans instead? See consumer pricing. Prefer pay-as-you-go? See platform.rivenai.io/pricing.
Plans
Every plan includes a monthly token allocation.
Pay-As-You-Go
- Pay only for what you use
- $3.00 per 1M tokens
- All Riven models
- No monthly minimum
- OpenAI-compatible API
Riven Pro
- Single seat
- All standard models
- Priority queue
- Memory
- OpenAI-compatible API
- 3M tokens/month included
- $5/mo bundled API credit
- 200 agentic steps/mo
- CLI + platform console access
Riven Team
- Multi-seat
- Per-seat billing
- All standard models
- Shared memory
- Admin console
- 4M tokens/month/seat included
- $10/seat/mo bundled API credit
- 1000 agentic steps/seat/mo
- Team admin + audit
How it works
How allocation works
No per-token invoices. Each plan grants a fixed pool of tokens (input + output combined) every month.
Free
100 requests per day across all Riven models — enough to explore and prototype.
Riven Pro
100,000 tokens per month plus a $5 bundled API credit for individual builders.
Pay-As-You-Go
$3.00 per 1M tokens, no monthly minimum — scale past any plan cap without upgrading.
Models
Every plan includes access to the full model catalog.
| Model | Provider | Context window | Capabilities |
|---|---|---|---|
| qwen2.5-7b-instruct | Ultra-native | 32,768 | chat, code |
| qwen3.6-35b-a3b-fp8 | Ultra-native | 131,072 | chat, vision, code, reasoning |
| nomic-embed-text | Ultra-native | 2,048 | embeddings |
| rvn-assistant-v2 | Ultra-native | 128,000 | chat, code, reasoning |
| gpt-5.5 | openai | 256,000 | chat, vision, code, reasoning |
| gpt-5.5-mini | openai | 256,000 | chat, vision, code, reasoning |
| gpt-5.5-nano | openai | 256,000 | chat, vision, code |
| gpt-5.4-mini | openai | 128,000 | chat, vision, code, reasoning |
| gpt-5.4-nano | openai | 128,000 | chat, vision, code |
| gpt-5 | openai | 128,000 | chat, vision, code, reasoning |
| claude-opus-4-8 | anthropic | 200,000 | chat, vision, code, reasoning |
| claude-opus-4-7 | anthropic | 200,000 | chat, vision, code, reasoning |
| claude-sonnet-4-6 | anthropic | 200,000 | chat, vision, code, reasoning |
| claude-haiku-4-5 | anthropic | 200,000 | chat |
| gemini-2.5-pro | 1,000,000 | chat, vision, code, reasoning | |
| gemini-2.5-flash | 1,000,000 | chat, vision, code | |
| gemini-2.0-pro | 2,000,000 | chat, vision, code, reasoning | |
| gemini-2.0-flash | 1,000,000 | chat, vision, code | |
| grok-3 | xai | 128,000 | chat, vision, code, reasoning |
| grok-3-mini | xai | 128,000 | chat, vision, code |
| grok-2 | xai | 128,000 | chat, vision, code, reasoning |
| mistral-large-latest | mistral | 256,000 | chat, vision, code, reasoning |
| mistral-medium-latest | mistral | 128,000 | chat, code |
| mistral-small-latest | mistral | 128,000 | chat, code |
| deepseek-v4 | deepseek | 128,000 | chat, vision, code, reasoning |
| deepseek-v4-lite | deepseek | 128,000 | chat, code |
| llama-4-maverick | meta | 256,000 | chat, vision, code, reasoning |
| llama-4-scout | meta | 128,000 | chat, vision, code |
| kimi-k2 | moonshot | 256,000 | chat, vision, code, reasoning |
| kimi-k2.5 | moonshot | 256,000 | chat, vision, code, reasoning |
| kimi-k1.5 | moonshot | 256,000 | chat, code, reasoning |
| command-a-03-2025 | cohere | 256,000 | chat, code, reasoning |
| command-r7b-12-2024 | cohere | 128,000 | chat, code |
| command-r-plus-08-2024 | cohere | 128,000 | chat, code, reasoning |
| sonar-reasoning-pro | perplexity | 128,000 | chat, code, reasoning |
| sonar-pro | perplexity | 128,000 | chat |
| sonar | perplexity | 128,000 | chat |
| together-llama-4-maverick | together | 256,000 | chat, vision, code, reasoning |
| together-qwen3-30b-a3b | together | 131,072 | chat, vision, code, reasoning |
| jamba-1.6-large | ai21 | 256,000 | chat, code, reasoning |
| qwen-max | alibaba | 128,000 | chat, vision, code, reasoning |
| qwen-plus | alibaba | 128,000 | chat, vision, code |
| qwen-turbo | alibaba | 128,000 | chat, vision, code |
What if I need more?
For pay-as-you-go pricing, higher limits, SSO, or on-premise deployment, use the Platform console — a separate PAYG + Enterprise surface.
- Pay-as-you-go token pricing
- SAML / Entra ID SSO
- RBAC and audit logs
- Private model deploys
- Dedicated support + SLA
- On-premise deployment
Questions
How does the monthly allocation work?
Each plan includes a fixed number of tokens (input + output combined) per calendar month. Your allocation resets automatically on the 1st of every month, UTC.
What happens if I use my full allocation?
API calls are paused until your next monthly reset, or you can upgrade to a higher tier instantly for more headroom.
Do unused tokens roll over?
No. Allocations reset to the plan amount every month — they do not accumulate or carry over.
Can I change plans anytime?
Yes. Upgrade or downgrade at any time from the console Plan page. Upgrades apply immediately.
Prefer pay-as-you-go?
For token-metered PAYG billing with no monthly commitment, use platform.rivenai.io instead of a dev subscription.