Guaranteed 30% off your current AI inference bill for teams spending $200 or more per month.
Book a call →
Flat monthly pricing for AI inference
One fixed monthly price. No billing surprises. No usage calculations.
Start free and scale when you're ready.
Free
For developers getting started with Oxlo.ai.
- 60 requests / day
- Requests may be queued behind paid plans
- Access to 12 plus open source models
- Clear usage limits
- No credit card required
Pro
For developers building and shipping AI-powered products.
- 300 requests / day
- Optimized models under 8B parameters
7 day free trial
- Faster request handling
- Access to optimised models for development and prototyping
- Higher throughput for development workloads
Premium
For teams running production workloads.
- 2,000 requests / day
- Production-grade performance
7 day free trial.
- Priority execution
- Higher and consistent throughput
- All large reasoning models including DeepSeek R1 and Kimi K2
FOR HIGH-VOLUME TEAMS
Enterprise
For teams ready to cut their AI infrastructure costs significantly.
OUR COMMITMENT
30% off your current AI bill.
Guaranteed. For teams spending $200 or more per month on AI inference with any provider.
No commitment. 30 minute conversation.
- Custom usage limits
- Dedicated support
- Tailored deployment options
Compare the plans
All plans use request-based pricing. No token calculations.
| Usage & Limits | ||||
|---|---|---|---|---|
| Requests included | 60 / day | 300 / day | High request limits | Custom |
| Burst rate limit | 5 / minute | 30 / min | 120 / min (tunable) | Custom |
| Monthly request cap | Yes | Yes | None | Custom |
| Request priority level | Lowest | Standard | High | Dedicated |
| Models & Performance | ||||
| Optimized models over 8B | No | Limited | Yes | Yes |
| Production-grade inference | No | No | Yes | Yes |
| Priority execution | Lowest | Medium | Highest | Optional |
| Average Response Latency | ≤ 7 seconds | ≤ 1 second | ≤ 100 ms | - tunable |
| Request & Context Limits (Caps are for safety and performance, not billing) | ||||
| Input tokens / request | Up to 8K | Up to 16K | Up to 32K | Custom (up to 128K) |
| Output tokens / request | Up to 2K | Up to 4K | Up to 8K | Custom (up to 128K) |
| Pricing & Billing | ||||
| Request-based pricing | Yes | Yes | Yes | Yes |
| Token-based billing | No | No | No | No |
| Fixed monthly limits | Yes | Yes | Yes | Custom |
| Usage limits visible upfront | Yes | Yes | Yes | Yes |
| Developer Experience | ||||
| Open-source models | Yes | Yes | Yes | Yes |
| Simple API integration | Yes | Yes | Yes | Yes |
| Model-agnostic pricing | Yes | Yes | Yes | Yes |
| Support level | Community | Community | Priority | Dedicated |
| Infrastructure & Technical Differentiation | ||||
| Gateway-level request metering | Yes | Yes | Yes | Yes |
| Pricing independent of prompt length | Yes | Yes | Yes | Yes |
| Traffic prioritization by plan | No | Yes | Yes | Yes |
| Async and batch-friendly workloads | Yes | Yes | Yes | Yes |
Pricing FAQ
With Oxlo.ai's request-based pricing, you pay a flat monthly subscription that includes a set number of API requests per day. Each request costs the same regardless of how many tokens are in your prompt or response. A 100-token prompt costs the same as a 50,000-token prompt. This is fundamentally different from token-based pricing used by OpenAI, Together AI, Fireworks AI, OpenRouter, and Replicate.
For teams running long-context or reasoning model workloads, yes. Together AI, Fireworks AI, and OpenRouter all charge per token, so costs scale linearly with prompt length. Running 500 API calls per day with 3,000-token prompts costs approximately $40-60/month on these providers vs $49.90/month on Oxlo.ai Premium. But as prompt length increases beyond 10,000 tokens, Oxlo.ai can be 10-100x cheaper since every request costs the same flat rate.
Yes. New users get a 7-day free trial with full access to all 40+ models including Qwen 3 32B, Llama 3.3 70B, DeepSeek R1, and premium image generation. No credit card required to start. The Free tier (60 requests/day, 16+ models) is available permanently.
When you reach your daily request limit, additional requests are queued until the next day or you can upgrade your plan for higher limits. There are no overage charges - your costs are always predictable and fixed. This is unlike token-based providers where a single runaway prompt can spike your bill.
Yes, you can upgrade or downgrade your plan at any time. When upgrading, you get immediate access to the higher plan's limits. All plans are billed monthly with no long-term contracts required.
Yes. Teams currently spending $200 or more per month on AI inference with providers like Together AI, Fireworks AI or OpenRouter are eligible for our Enterprise plan which guarantees a minimum 30 percent reduction on their current monthly bill. Contact us at hello@oxlo.ai to discuss your current usage.