Telluvian vs OpenRouter: LLM routing compared
Telluvian and OpenRouter both put many models behind one API, but they choose models differently. Telluvian's Model Select predicts how well each of 190+ models will answer a given prompt and sends it to the cheapest one that clears your quality bar, with optional per-token hallucination scores. OpenRouter offers a larger catalogue of 500+ models; you choose the model yourself or use its Auto Router, which ranks models by what OpenRouter users spend on for that type of task.
Last reviewed
At a glance
| Telluvian | OpenRouter | |
|---|---|---|
| Model selection method | Prompt-aware. Model Select predicts each model’s performance on the specific prompt and picks the cheapest one that clears your xPerf quality bar. | You name the model, or the Auto Router classifies the prompt by task type and ranks models by aggregate user spend on that task over a trailing 7 days. Providers for a model are picked by stability, then weighted towards the cheapest. |
| Models and providers | 190+ models from OpenAI, Anthropic, Google, Qwen, DeepSeek, xAI, Z-AI, Moonshot and Meta, behind one OpenAI-compatible API. | 500+ models from 80+ providers on paid plans. |
| Pricing model | Prepaid, pay as you go. Provider list price for tokens, plus $0.05 per 1M input tokens for Model Select and $1.00 per 1M completion tokens for hallucination scores (optional). | Provider prices with no markup on tokens. A platform fee on credit purchases: 5.5% on Standard, 8% on Business. Bring your own key is free up to $25,000 a month of list price, then 5%. No fee for the Auto Router. |
| Hallucination detection | Built in. Per-token hallucination scores from probes reading an open-weight proxy model, including for closed models. Off by default. | None documented. |
Key differences
Model choice
Telluvian predicts each model's performance on the specific prompt; OpenRouter's Auto Router ranks models by aggregate user spend on that task type over the past 7 days.
Quality bar
Telluvian takes one per request with xPerf, a model name or a number on a fixed scale that keeps its meaning as new models launch.
Hallucination detection
Telluvian can return a score for every generated token; OpenRouter does not document an equivalent.
Catalogue
OpenRouter lists 500+ models, including free ones; Telluvian selects among 190+ from the major providers.
Fees
OpenRouter charges 5.5% on credit purchases (8% on Business); Telluvian has no top-up fee and charges $0.05 per 1M input tokens when it picks the model.
Frequently asked questions
Is Telluvian an alternative to OpenRouter?
Yes, for chat workloads. Telluvian gives you one OpenAI-compatible API for 190+ models from OpenAI, Anthropic, Google, Qwen, DeepSeek, xAI and others, and also chooses which of them should answer each prompt.
Which is cheaper, Telluvian or OpenRouter?
Both charge model tokens at provider list price. OpenRouter adds a fee when you buy credits (5.5% on Standard, 8% on Business). Telluvian has no top-up fee and adds $0.05 per 1M input tokens when it picks the model. In practice, which model answers tends to matter more to the bill than either fee.
Does OpenRouter choose the model automatically?
Yes, through its Auto Router, which classifies your prompt by task type and ranks models by what OpenRouter users have spent on for that kind of task over the past 7 days. Telluvian's Model Select instead predicts how well each candidate will answer the specific prompt and picks the cheapest one that meets your xPerf bar.
Does OpenRouter detect hallucinations?
We could not find hallucination detection in OpenRouter's public docs. Telluvian returns a hallucination score for every generated token when you set include_scores to true, for any model it serves.
How do I move from OpenRouter to Telluvian?
Telluvian has an OpenAI-compatible API. Point your OpenAI SDK at https://api.telluvian.ai/v1 with a Telluvian key and send telluvian/gallery-1 as the model. To keep requests with providers you have agreements with, limit the candidates with modelZoo, for example "anthropic/*,openai/*". Tool calling is coming soon.
Try it on your own prompts
Point an OpenAI SDK at Telluvian, send telluvian/gallery-1 as the model, and see which model answers each request. Questions about your use case go straight to the team.
Sources
Competitor details come from their own public docs and pricing pages, checked on . Where a page did not say, neither do we. Spotted something out of date? Tell us at hello@telluvian.ai.