haystack-experimental
Experimental components and features for the Haystack LLM framework.
Decision gist · record as of 2026-08-14
No, unless you have legacy code that depends on a discontinued experiment and cannot migrate to haystack-ai. The package is archived, abandoned, and will not be maintained. Any new Haystack work should use haystack-ai directly. If you do install, pin to 0.19.0.post1 explicitly and understand that compatibility with future haystack-ai releases is not guaranteed.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >=3.10.
- Most experiments have graduated to haystack-ai; check the Haystack documentation first.
- Pinning to 0.19.0.post1 is mandatory—no future releases will be published.
License · maintenance · safety
Apache-2.0 (permissive) — Licensed under Apache-2.0 (permissive). No licensing barrier to use, but the abandoned status means you assume all maintenance and compatibility risk yourself.
last release 2026-07-28 (17 days) · last repo commit 2026-07-29 · 65 stars · archived
0 known vulnerabilities (OSV.dev, 2026-08-14) · 713,714 downloads/mo, #5,250 on PyPI
Alternatives
Verify before relying
pip install "haystack-experimental==0.19.0.post1"
from haystack_experimental import SomeExperiment
# Use only if the experiment is not available in haystack-ai- Exact compatibility window with haystack-ai versions beyond February 2026 is unclear; drift risk increases over time.
- Whether all discontinued experiments remain functional or have bitrotted is not specified in the fact sheet.
What it is and what it does
haystack-experimental was a staging ground for early-access features in the Haystack LLM framework, allowing users to test and provide feedback on components before they were committed to the main release. The package has now been archived and is no longer maintained. Most experiments—including Agent, AsyncPipeline, Tool calling, multimodality support, and Human-in-the-Loop features—have graduated into haystack-ai itself. A smaller set of experiments were discontinued entirely and exist only in this final release.
The 0.19.0.post1 release is frozen and will receive no updates, bug fixes, or compatibility patches. It depends only on haystack-ai and supports Python 3.10 through 3.13. If you need an experiment that graduated, use haystack-ai instead. If you depend on a discontinued experiment, you must pin this package explicitly and accept that it will drift out of sync with newer haystack-ai releases over time.
Use it for
- Accessing discontinued experiments (e.g., OpenAIFunctionCaller, OpenAPITool) that were not adopted into Haystack core and exist nowhere else.
- Maintaining legacy code that was written against a specific experimental feature before it graduated or was discontinued.
- Pinning a known-working snapshot of an experiment for production use when the graduated version in haystack-ai has breaking changes.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
No, unless you have legacy code that depends on a discontinued experiment and cannot migrate to haystack-ai.
The package is archived, abandoned, and will not be maintained. Any new Haystack work should use haystack-ai directly. If you do install, pin to 0.19.0.post1 explicitly and understand that compatibility with future haystack-ai releases is not guaranteed.
Install
haystack-experimental on PyPI
Before you install
The package is archived and abandoned as of the final release on 2026-07-28. Install friction is low (pure Python wheel), but maintenance status is critical: no further updates, bug fixes, or compatibility work will occur. Only install if pinning to 0.19.0.post1 explicitly for a discontinued experiment not available elsewhere.
Requires Python >=3.10. Most experiments have graduated to haystack-ai; check the Haystack documentation first. Pinning to 0.19.0.post1 is mandatory—no future releases will be published.
License in practice
Licensed under Apache-2.0 (permissive). No licensing barrier to use, but the abandoned status means you assume all maintenance and compatibility risk yourself.
Quickstart
pip install "haystack-experimental==0.19.0.post1"
from haystack_experimental import SomeExperiment
# Use only if the experiment is not available in haystack-aiVerify before relying
- Exact compatibility window with haystack-ai versions beyond February 2026 is unclear; drift risk increases over time.
- Whether all discontinued experiments remain functional or have bitrotted is not specified in the fact sheet.
Package facts
| License | Apache-2.0 permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 1 packagehaystack-ai |
| Maintenance | Abandoned 17 days since the last release |
| Last repo commit | repository archived |
| First released | |
| Downloads | 713,714 / month, #5,250 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: Science/ResearchLicense :: Freely DistributableLicense :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: PythonProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Topic :: Scientific/Engineering :: Artificial Intelligence |
Evidence: haystack_experimental-0.19.0.post1-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “haystack experimental features”
- haystack-experimentalProvides archived experimental features for Haystack LLM framework…
- openinference-instrumentation-haystackAuto-instruments Haystack LLM pipelines and components to emit…
- opentelemetry-instrumentation-haystackAdds distributed tracing to Haystack LLM applications by…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also farm-haystack · haystack-ai · isa-rwval · openinference-instrumentation-haystack · opentelemetry-instrumentation-haystack · django-haystack · strands-agents-evals · sagemaker-experiments · llama-index-cli · qiskit-experiments