Migrate from honeyhive-bundled to v1

honeyhive-bundled was a temporary distribution of the Python SDK with pre-release packaging. The stable v1 SDK is published as honeyhive. Most imports stay the same because both packages expose the honeyhive Python module. The migration is mainly a dependency change plus cleanup of pre-release patterns.

Recent honeyhive-bundled releases and honeyhive v1 share the same top-level import style: from honeyhive import HoneyHiveTracer, trace, evaluate.

Use a coding agent to migrate

Use this HoneyHive migration guide: https://docs.honeyhive.ai/v2/sdk-reference/python/migration/honeyhive-bundled-to-v1.md

Explore my codebase to find where HoneyHive Python SDK dependencies and tracing setup are configured. Ask me any questions needed to understand the project context, runtime, and migration constraints. Then present a concise migration plan for my confirmation. After I confirm the plan, implement the migration changes.

What changes

Area	`honeyhive-bundled`	`honeyhive>=1.0.0`
Package name	`honeyhive-bundled`	`honeyhive`
Release channel	Pre-release / bundled distribution	Stable package
Default API URL	`https://api.honeyhive.ai`	`https://api.dp1.us.honeyhive.ai`
Project routing	Older examples may pass `project=`	`project=` still accepted but emits a `DeprecationWarning` and will be removed in v2
`honeyhive.config`	Alias for the `TracerConfig` class (callable)	Submodule containing `TracerConfig`, `SessionConfig`, `EvaluationConfig`, etc.
`clear_baggage_context`	Exported from `honeyhive`	Removed from the public API
`honeyhive` CLI command	Registered as a console script	HoneyHive CLI is now installed separately, see HoneyHive CLI
OpenTelemetry minimum	`>=1.20.0`	`>=1.41.0`

If your deployment relies on the default API host (for example, when running through a corporate proxy or allow-list), set HH_API_URL to your existing host before upgrading. v1 changes the default host to https://api.dp1.us.honeyhive.ai.

1. Replace the dependency

pip uninstall honeyhive-bundled -y
pip install "honeyhive>=1.0.0"

If you use extras, install the matching extras on honeyhive:

pip install "honeyhive[openinference-openai]"
pip install "honeyhive[openinference-anthropic]"
pip install "honeyhive[openinference-langchain]"

pip install "honeyhive[traceloop-openai]"
pip install "honeyhive[traceloop-anthropic]"
pip install "honeyhive[traceloop-aws-bedrock]"

honeyhive>=1.0.0 keeps every OpenInference and Traceloop extra that honeyhive-bundled shipped and adds new ones: LangChain, LiteLLM, OpenAI Agents, Claude Agent SDK, AWS Strands, and the openinference-llm-providers / traceloop-llm-providers multi-provider bundles. Run pip check after upgrading if your environment pins OpenTelemetry packages, because honeyhive v1 requires OpenTelemetry >=1.41.0.

The Google AI extra is renamed under the hood: bundled installed openinference-instrumentation-google-generativeai, but honeyhive[openinference-google-ai]>=1.0.0 installs openinference-instrumentation-google-genai instead. If you import the instrumentor directly, update the import path.

Verify the install:

import honeyhive

print(honeyhive.__version__)

honeyhive-bundled reports its internal honeyhive.__version__ as 1.0.0rc9.post2 even at PyPI version 1.1.1. After migrating, honeyhive.__version__ will report the published v1 version (for example 1.0.2).

2. Update imports

A few symbols that honeyhive-bundled exported are no longer part of the public API in v1. Update or remove these imports:

Import in `honeyhive-bundled`	Status in `honeyhive>=1.0.0`	What to do
`from honeyhive import clear_baggage_context`	Removed	Delete the call. Baggage is managed automatically by `HoneyHiveTracer` during init and session creation.
`honeyhive.config(...)` (config alias for `TracerConfig`)	`honeyhive.config` is now a submodule, not a class	Import the class explicitly: `from honeyhive.config import TracerConfig`.
`honeyhive.__version_bundled__`	Removed	Read `honeyhive.__version__` instead.

Everything else keeps the same top-level import style:

from honeyhive import HoneyHiveTracer, trace, evaluate

If your dependency file names the bundled package, update it:

[project]
dependencies = [
    "honeyhive-bundled[openinference-openai]>=1.0.1",
]

[project]
dependencies = [
    "honeyhive[openinference-openai]>=1.0.0",
]

The honeyhive console-script that honeyhive-bundled registered is not published by honeyhive>=1.0.0, and the underlying honeyhive.cli module is deprecated. Replace those calls with the official HoneyHive CLI (@honeyhive/cli), which is a separate package that maps one-to-one to the HoneyHive REST API.

3. Initialize instrumentors separately

HoneyHiveTracer.init() does not consume an instrumentors= argument in either honeyhive-bundled or honeyhive>=1.0.0. Bundled accepted it silently via **kwargs and never called instrumentor.instrument(), so spans were never produced from that list. Initialize provider instrumentors yourself after HoneyHiveTracer.init() and pass tracer.provider so they emit spans into the HoneyHive tracer.

from honeyhive import HoneyHiveTracer
from openinference.instrumentation.openai import OpenAIInstrumentor

tracer = HoneyHiveTracer.init(
    api_key="your-api-key",
    project="support-bot",
    instrumentors=[OpenAIInstrumentor()],  # silently ignored
)

from honeyhive import HoneyHiveTracer
from openinference.instrumentation.openai import OpenAIInstrumentor

tracer = HoneyHiveTracer.init(api_key="your-api-key")
OpenAIInstrumentor().instrument(tracer_provider=tracer.provider)

Initialize the tracer before any instrumentor. Instrumentors need the HoneyHive tracer provider to route spans correctly.

evaluate() is different: it still accepts an instrumentors= keyword and uses the factories per datapoint. See section 6.

See OpenAI, Anthropic, and LangChain for provider-specific setup.

4. Remove project routing from new code

v1 accepts project= for compatibility but emits a DeprecationWarning and will remove it in v2. Project context is inferred from the API key. honeyhive-bundled accepted project= silently, so after migrating, expect a warning on every initialization until you drop the argument.

DeprecationWarning: The 'project' argument to HoneyHiveTracer.init() is
deprecated; it will be removed in v2.0. Remove it from HoneyHiveTracer.init()
calls — the backend infers project context from the API key.

import os
from honeyhive import HoneyHiveTracer

tracer = HoneyHiveTracer.init(
    api_key=os.environ["HH_API_KEY"],
    project=os.environ["HH_PROJECT"],
    source="production",
)

import os
from honeyhive import HoneyHiveTracer

tracer = HoneyHiveTracer.init(
    api_key=os.environ["HH_API_KEY"],
    source="production",
)

For existing v1 deployments, you can remove project= after confirming your API keys are scoped to the intended project.

5. Prefer tracer instance methods for enrichment

Free helpers such as enrich_span() and enrich_session() are still exported and still work in v1, but tracer instance methods are the primary pattern because they are explicit and behave correctly when multiple tracers are in play.

from honeyhive import enrich_span, trace

@trace(event_type="tool")
def retrieve(query):
    docs = search(query)
    enrich_span(metadata={"query": query}, outputs={"docs": docs})
    return docs

from honeyhive import HoneyHiveTracer, trace

tracer = HoneyHiveTracer.init(api_key="your-api-key")

@trace(event_type="tool", tracer=tracer)
def retrieve(query):
    docs = search(query)
    tracer.enrich_span(metadata={"query": query}, outputs={"docs": docs})
    return docs

See Enriching Traces.

6. Check `evaluate()` usage

The stable v1 experiment API expects datasets with ground_truth and functions that receive the full datapoint.

from honeyhive import evaluate
from openinference.instrumentation.openai import OpenAIInstrumentor

dataset = [
    {
        "inputs": {"question": "What is the capital of France?"},
        "ground_truth": {"answer": "Paris"},
    },
]

def run_pipeline(datapoint):
    question = datapoint["inputs"]["question"]
    return {"answer": answer(question)}

def exact_match(outputs, inputs, ground_truth):
    return {"score": outputs["answer"] == ground_truth["answer"]}

evaluate(
    function=run_pipeline,
    dataset=dataset,
    evaluators=[exact_match],
    api_key="your-api-key",
    instrumentors=[lambda: OpenAIInstrumentor()],
)

Use instrumentor factories in evaluate(), for example lambda: OpenAIInstrumentor(). The SDK creates a fresh tracer per datapoint and initializes instrumentors inside the evaluation runner, so pass a factory instead of a shared pre-initialized instrumentor.

7. Validate the migration

Run:

pip check
PYTHONWARNINGS=default python your_app.py

Confirm:

pip show honeyhive-bundled no longer finds an installed package
pip show honeyhive shows version 1.0.0 or later
No from honeyhive import clear_baggage_context imports remain
Any honeyhive.config(...) callers use from honeyhive.config import TracerConfig
No HoneyHiveTracer.init(..., instrumentors=...) calls remain
HH_API_URL is set if you rely on a non-default host
Traces appear in the expected project
Provider spans appear after instrumentor initialization
Experiments show ground_truth and evaluator scores

Migration checklist

Migrate to SDK v1

Upgrade older SDK code to stable v1 semantics.

Migrate from Logger to v1

Replace honeyhive-logger calls with SDK v1 tracing.

SDK Overview

Choose the right SDK and integration path.

Integrations

Install and initialize provider instrumentors.

Environment Variables

Configure SDK runtime behavior.

Python SDK

TypeScript API SDK

TypeScript Control Plane SDK

Semantic Conventions

OpenAPI Client

Migrate from honeyhive-bundled to v1

What changes

1. Replace the dependency

2. Update imports

3. Initialize instrumentors separately

4. Remove project routing from new code

5. Prefer tracer instance methods for enrichment

6. Check `evaluate()` usage

7. Validate the migration

Migration checklist

Migrate to SDK v1

Migrate from Logger to v1

SDK Overview

Integrations

Environment Variables

​What changes

​1. Replace the dependency

​2. Update imports

​3. Initialize instrumentors separately

​4. Remove project routing from new code

​5. Prefer tracer instance methods for enrichment

​6. Check evaluate() usage

​7. Validate the migration

​Migration checklist

​Related

Migrate to SDK v1

Migrate from Logger to v1

SDK Overview

Integrations

Environment Variables

What changes

1. Replace the dependency

2. Update imports

3. Initialize instrumentors separately

4. Remove project routing from new code

5. Prefer tracer instance methods for enrichment

6. Check `evaluate()` usage

7. Validate the migration

Migration checklist

Related