mem0ai/mem0 · error · ValueError

Invalid compression_type: {values['compression_type']}. Must

Error message

Invalid compression_type: {values['compression_type']}. Must be one of: {', '.join(valid_types)}, or None

What it means

Within the Azure AI Search config's pre-validator, compression_type (when not None) must be exactly 'scalar' or 'binary', compared case-insensitively via .lower(). Any other string raises ValueError naming the invalid value and the allowed options. This runs before extra-field checks return, at config construction time.

Source

Thrown at mem0/configs/vector_stores/azure_ai_search.py:50

        # Check for use_compression to provide a helpful error
        if "use_compression" in extra_fields:
            raise ValueError(
                "The parameter 'use_compression' is no longer supported. "
                "Please use 'compression_type=\"scalar\"' instead of 'use_compression=True' "
                "or 'compression_type=None' instead of 'use_compression=False'."
            )

        if extra_fields:
            raise ValueError(
                f"Extra fields not allowed: {', '.join(extra_fields)}. "
                f"Please input only the following fields: {', '.join(allowed_fields)}"
            )

        # Validate compression_type values
        if "compression_type" in values and values["compression_type"] is not None:
            valid_types = ["scalar", "binary"]
            if values["compression_type"].lower() not in valid_types:
                raise ValueError(
                    f"Invalid compression_type: {values['compression_type']}. "
                    f"Must be one of: {', '.join(valid_types)}, or None"
                )

        return values

    model_config = ConfigDict(arbitrary_types_allowed=True)

View on GitHub (pinned to 001c235229)

Solutions

Use compression_type="scalar" or "binary" (any casing), or None to disable compression
In YAML, write compression_type: (empty) or null, never the string "none"
Strip whitespace from values injected from external config sources

Example fix

# before
"config": {"service_name": S, "api_key": K, "compression_type": "none"}

# after
"config": {"service_name": S, "api_key": K, "compression_type": None}

Defensive patterns

Strategy: validation

Validate before calling

VALID = {"scalar", "binary"}
if "compression_type" in cfg and cfg["compression_type"] is not None:
    if str(cfg["compression_type"]).strip().lower() not in VALID:
        raise ConfigError(f"compression_type must be one of {sorted(VALID)} or None")

Type guard

from typing import Optional

def is_valid_compression(v) -> bool:
    return v is None or str(v).strip().lower() in {"scalar", "binary"}

Try / catch

try:
    Memory.from_config(config)
except ValueError as e:
    if "Invalid compression_type" in str(e):
        config["vector_store"]["config"]["compression_type"] = None
        Memory.from_config(config)
    else:
        raise

Prevention

Use an enum/Literal for compression_type in your own config schema; use None for 'off'
Beware YAML quoting: 'null' or 'none' strings are not None

When it happens

Trigger: Setting compression_type to e.g. 'none', 'int8', 'pq', 'product', or 'SCALAR ' (trailing space) in the azure_ai_search config; note None disables compression and is valid.

Common situations: Mapping Azure terminology (e.g. 'rescoringOptions', 'half', 'int8') into compression_type; users assuming an 'off' string instead of None; values sourced from YAML where quotes make "null" a string.

Related errors

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/0bca913d21f9e3a9. Report an issue: GitHub.