#3333rd of 42 in Gateways and observability
coze-loop
Prompt playground, evaluation and tracing platform for developing and monitoring AI agents
- Stars
- 5.8k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Jan 2026
- Language
- Go
5,325 on 10 Apr 20265,799 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Coze Loop is the open-source edition of ByteDance's agent development and operations platform. It covers prompt authoring with a playground and version management, evaluation sets, evaluators and experiments, and trace reporting through SDKs in three languages. Models are connected through the Eino framework, with OpenAI and Volcengine Ark named in the README.
Who it is for: Teams building LLM agents who need prompt versioning, evals and tracing
Strengths
- Combines prompt versioning, evaluation experiments and trace observation in one deployable service
- Side-by-side playground comparison of output from different LLMs
- Deploys via Docker Compose (make compose-up) or a Helm chart for Kubernetes
- SDK works with both the commercial and open-source editions
Weaknesses
- Open-source edition exposes only core modules of the commercial product
- README setup examples centre on Volcengine Ark; other providers are documented only in the wiki
- README warns of SSRF, open registration and some privilege-escalation risks on public networks
What it needs
- no GPU
- Needs Docker Engine, Docker Compose, Kubernetes and Helm (optional), LLM provider API key
- Models: OpenAI, Volcengine Ark, BytePlus ModelArk, Models supported via Eino
- port 8082
Quick answers
Is coze-loop open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run coze-loop with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does coze-loop need a GPU?
No, it runs without one.
Also in Gateways and observability
See all 42| Rank | Project | Score |
|---|---|---|
| 1 | OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers | 96 out of 100 |
| 2 | new-apiGateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs | 90 out of 100 |
| 3 | MLflowTracing, evals, prompt registry and AI gateway plus classic ML tracking | 89 out of 100 |
| #4 | LiteLLMOpenAI-format gateway and Python SDK for calling 100+ LLM providers | 87 out of 100 |
| #5 | freellmapiOpenAI-compatible gateway pooling free-tier LLM providers with automatic failover | 86 out of 100 |