Self-hosted and open source - verified
Bifrost (Maxim)
Bifrost is listed under self-hosted and open source. Its first listed fee, in its own words: open source (Apache 2.0), self-hosted software; enterprise features listed separately. It states OpenAI-compatible and Anthropic-compatible requests. Routing stated: fallbacks and load balancing. Read from the vendor's own pages on October 18, 2026.
Read from docs.getbifrost.ai, raw.githubusercontent.com, github.com. Docs - Compare every gateway
Status
- Status, as we read it
available; open source and enterprise editionsinferred
Our reading: The page implies this status without labelling it.
“Deploy in seconds with zero configuration”
docs.getbifrost.ai
How Bifrost charges
Every fee line Bifrost states, in its own words and units, with the quote. Nothing is converted.
| Product | Fee, in the vendor's words | Applies to | Quote and source |
|---|---|---|---|
| Bifrost (Maxim) | open source (Apache 2.0) | self-hosted software; enterprise features listed separately | “Apache License” |
Access and keys
- Free tier
- Not stated
Open-source software; free tier not applicable.
- Bring your own key
yes (your provider API keys, weighted across keys)inferred
Our reading: Read as bring your own key from distributing requests across your own provider API keys.
“Intelligent API key distribution with weighted load balancing, model-specific filtering, and automatic failover.”
docs.getbifrost.ai - BYOK fee
- Not stated
Not applicable.
API formats
- OpenAI-compatible
yes
“through a single OpenAI-compatible API”
raw.githubusercontent.com - Anthropic-compatible
yes (drop-in for Anthropic SDKs)
“Drop-in replacement for the Anthropic Python and TypeScript SDKs.”
docs.getbifrost.ai - OpenAI base URL
http://localhost:8080/openai (self-hosted default)
“base_url = "http://localhost:8080/openai"”
raw.githubusercontent.com - Anthropic base URL
http://localhost:8080/anthropic (self-hosted default)
“base_url = "http://localhost:8080/anthropic"”
raw.githubusercontent.com
Routing
- Fallbacks
yes
“When your primary provider fails, Bifrost switches to backups automatically.”
docs.getbifrost.ai - Load balancing
yes (weighted)
“Intelligent API key distribution with weighted load balancing, model-specific filtering, and automatic failover.”
docs.getbifrost.ai - Automatic model routing
- Not stated
Enterprise "Adaptive Routing" picks the healthiest provider and key; automatic model selection not stated.
Caching
- Exact-match cache
- Not stated
Not stated on the checked pages.
- Semantic cache
yes
“Intelligent response caching based on semantic similarity.”
docs.getbifrost.ai
Controls and observability
- Spend controls
hierarchical budgets
“Hierarchical cost control with budgets and rate limits at virtual key, team, and customer levels.”
docs.getbifrost.ai - Rate limits
yes (virtual key, team, customer)
“Hierarchical cost control with budgets and rate limits at virtual key, team, and customer levels.”
docs.getbifrost.ai - Guardrails
- Not stated
Not stated on the checked pages.
- Observability
built-in observability, Prometheus metrics
“Monitor every AI request in real-time.”
docs.getbifrost.ai - Telemetry and log export
OpenTelemetry (OTLP); log exports on Enterprise
“OTLP integration for distributed tracing with Grafana, New Relic, Honeycomb, and more.”
docs.getbifrost.ai
Data, deployment and license
- Data logging and retention
- Not stated
Self-hosted; not stated.
- Regions
- Not stated
Self-hosted; in-VPC deployments listed under Enterprise.
- Self-hosting
yes (npx or Docker)
“npx -y @maximhq/bifrost”
raw.githubusercontent.com - License
Apache 2.0
“Version 2.0, January 2004”
raw.githubusercontent.com
Vendor-stated counts
Model and provider counts as Bifrost states them, dated. When two pages disagree, both are shown.
20+ providers (docs) (vendor-stated, Oct 18 2026)
“unifies access to 20+ providers”
docs.getbifrost.ai 23+ providers (README) (vendor-stated, Oct 18 2026)
“unifies access to 23+ providers”
raw.githubusercontent.com
Research notes
Docs (20+) and README (23+) state different provider counts; both recorded.
FAQ
How much does Bifrost cost?
In Bifrost's own terms: open source (Apache 2.0) (self-hosted software; enterprise features listed separately). As of October 18, 2026; fees are not converted.
Does Bifrost support bring your own key?
Yes (your provider API keys, weighted across keys) (our reading of the quote), per the vendor: "Intelligent API key distribution with weighted load balancing, model-specific filtering, and automatic failover.". BYOK fee: not stated. Not applicable.
Is Bifrost OpenAI-compatible or Anthropic-compatible?
OpenAI: yes, base URL http://localhost:8080/openai. Anthropic: yes (drop-in for Anthropic SDKs), base URL http://localhost:8080/anthropic.
Sources
Last verified from each vendor's own docs, pricing pages, READMEs and license files; every quote was string-matched against the page that day. Fees are shown in the vendor's own words and change without notice, so check the linked source before you commit.
Newsletter
Know when a gateway changes its fees
One email per week. New model releases, routing and pricing changes and what they mean for what you build.