Fuelline is an OpenAI-compatible API for the best open models. One endpoint, one key — every request routed for speed and uptime, at some of the lowest per-token prices on the market.
Why Fuelline
Every request is scored for price, latency and health across our serving infrastructure. You send one call — Fuelline picks the fastest healthy route and falls back instantly when something breaks.
Errors, timeouts and rate limits are retried on the next healthy route — before the first token reaches you. You never see a dead endpoint.
Every request is weighed by cost, latency and live error rates — you get the fastest healthy route by default, without thinking about it.
Conversations stay pinned to the same serving node via x-session-id or prompt_cache_key — preserving prompt caches and cutting repeat costs.
/v1/chat/completions, /v1/models, streaming SSE, tools. Point any OpenAI SDK at the base URL — no new client code, no vendor lock-in.
Serving capacity is probed every 5 minutes; error streaks trigger an instant cooldown. The catalog and prices refresh themselves — no manual upkeep.
Per-token prices for every model, published live on this page. What you see in the catalog is what you pay — no hidden tiers, no surprises.
Catalog
Live catalog. Prices are per million tokens, in USD — what you see is what you pay.
Loading catalog…
| Model | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| Loading catalog… | |||
Integrate
Already using the OpenAI SDK? Change the base URL and key. That's the whole migration.
import os
from openai import OpenAI
client = OpenAI(
base_url="https://fuelline.dev/v1",
api_key=os.environ["FUELLINE_API_KEY"],
)
response = client.chat.completions.create(
model="zai-org/GLM-5.3-Flash",
messages=[{"role": "user", "content": "Hello!"}],
)