v1.4.7

npx -y @maximhq/bifrost --transport-version v1.4.7

docker pull maximhq/bifrost:v1.4.7
docker run -p 8080:8080 maximhq/bifrost:v1.4.7

Bifrost(HTTP)

1.4.7

feat: added multi level routing support for routing rules + vk based provider routing
feat: added support for filtering /v1/models responses based on virtual key configurations in the governance plugin
feat: add key-level model discovery status tracking to improve visibility into API key health and model availability.
feat: add routing engine decision logs
feat: view/hide toggle for password on login field @Aaryan Meena
fix: add custom model support to routing rules UI
fix: add passthrough for count tokens for anthropic integration
fix: add object field to responses response
fix: reduce provider API connection max idle time to 30 seconds to avoid ttl mismatch
fix: add message sanitization for huggingface provider to remove unsupported fields from chat messages
fix: use status instead of error field for tool result errors for anthropic provider @Edward Upton
chore: upgrade core to 1.4.3 and framework to 1.2.21
chore: fix CVE-2025-68121 by upgrading to Go 1.26.0
refactor: ListModelsRequest to use the common request handling pipeline instead of its own implementation
fix: add skip plugin pipeline flag to context for list models requests to avoid internal list models logs
fix: duplicate ‘not found’ error on custom provider deletion

Core

1.4.3

feat: added support for filtering /v1/models responses based on virtual key configurations in the governance plugin
feat: add routing engine decision logs to context
feat: add skip plugin pipeline flag to context
fix: add object field to responses response
fix: reduce provider API connection max idle time to 30 seconds to avoid ttl mismatch
fix: add message sanitization for huggingface provider to remove unsupported fields from chat messages
fix: use status instead of error field for tool result errors for anthropic provider @Edward Upton
chore: fix CVE-2025-68121 by upgrading to Go 1.26.0
refactor: ListModelsRequest to use the common request handling pipeline instead of its own implementation

Framework

1.2.21

feat: rename routing_engine_used column to routing_engines_used for multi-engine tracking (parsed as comma-separated string)
feat: add routing engine decision logs to logs table
chore: upgrade core to 1.4.3
chore: fix CVE-2025-68121 by upgrading to Go 1.26.0

governance

1.4.22

feat: added multi level routing support for routing rules + vk based provider routing
feat: cross-provider model matching - budget/rate-limit configs for gpt-4o now apply to openai/gpt-4o, gpt-4o-2024-08-06, etc.
feat: expand GovernanceData with ModelConfigs and Providers for in-memory reads
feat: added routing rules for dynamic routing of requests based on predefined rules
feat: add model and provider level governance - set budgets and rate limits on specific models or providers independent of virtual keys
feat: plugin now filters /v1/models responses based on virtual key configurations
chore: upgrade core to 1.4.1 and framework to 1.2.19
feat: add routing engine decision logs to context
chore: upgrade core to 1.4.3 and framework to 1.2.21
chore: fix CVE-2025-68121 by upgrading to Go 1.26.0

jsonparser

1.4.21

litellmcompat

0.0.11

logging

1.4.21

maxim

1.5.20

mocker

1.4.21

otel

1.1.21

semanticcache

1.4.20

telemetry

1.4.22

feat: support multiple routing engines in telemetry metrics with comma-separated label format
chore: upgrade core to 1.4.3 and framework to 1.2.21
chore: fix CVE-2025-68121 by upgrading to Go 1.26.0