Telluvian vs Portkey: AI gateway and routing compared
Telluvian and Portkey are both gateways in front of many models, built for different jobs. Telluvian chooses the model for each prompt by predicting which one will answer it well for the least money, and can score answers for hallucinations. Portkey focuses on operating the traffic: logging, prompt management, guardrails and routing by rules you define over metadata and request parameters.
Last reviewed
At a glance
| Telluvian | Portkey | |
|---|---|---|
| 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. | Rules you configure: conditional routing on metadata, request parameters or URL path, plus fallbacks and load balancing. |
| Models and providers | 190+ models from OpenAI, Anthropic, Google, Qwen, DeepSeek, xAI, Z-AI, Moonshot and Meta, behind one OpenAI-compatible API. | 200+ LLMs across every major provider; its open-source gateway lists 1,600+ LLMs from 45+ providers. |
| 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). | Free Developer tier with 10k recorded logs a month. Production is $49 a month for 100k logs, then $9 per extra 100k. Enterprise is custom. Priced on logs, not tokens. |
| Hallucination detection | Built in. Per-token hallucination scores from probes reading an open-weight proxy model, including for closed models. Off by default. | Guardrails include deterministic checks, LLM-based checks such as prompt injection, and partner integrations. Hallucination detection is not listed. |
Key differences
Model choice
Telluvian decides per prompt from predicted answer quality; Portkey follows conditions you write, with fallbacks and load balancing.
Quality bar
Telluvian's xPerf is a fixed scale, so a setting keeps its meaning as new models launch.
Hallucination detection
Telluvian can score every generated token; Portkey's guardrail docs do not list hallucination detection.
Pricing
Portkey charges a platform fee by recorded logs (a free tier, then $49 a month for 100k logs); Telluvian charges per token, with no subscription.
Frequently asked questions
Does Portkey choose the model automatically?
Not from the prompt. Portkey's conditional routing sends requests to targets based on rules you define over metadata, request parameters and URL path, with fallbacks and load balancing. Telluvian's Model Select predicts how well each model will answer each prompt and picks the cheapest one above your quality bar.
How does Portkey pricing compare with Telluvian?
Portkey charges for the platform by recorded logs: a free tier, then $49 a month for 100k logs on Production, plus overages. Telluvian has no subscription. You pay model tokens at provider list price, $0.05 per 1M input tokens when Model Select picks the model, and $1.00 per 1M completion tokens if you switch on hallucination scores.
Can Portkey detect hallucinations?
Its guardrail docs list deterministic checks, LLM-based checks and partner guardrails, but not hallucination detection. Telluvian scores every generated token for hallucination when you set include_scores to true.
Does Telluvian support tool calling?
Tool calling is coming soon. Today Telluvian serves chat through Chat Completions and the Responses API.
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.