Immutable solution version
Vertex AI Vector Search: list()/get_all fails on non-768-dimension indexes
Complete solution
the mem0 Vertex AI Vector Search integration builds the zero query vector with a hardcoded dimension of 768 and the config has no dimension field, so any index with a different dimension (OpenAI 1536/3072) rejects the query.
When it applies
Not specified by the original author.
Instructions
[mem0ai/mem0#6419]: the mem0 Vertex AI Vector Search integration builds the zero query vector with a hardcoded dimension of 768 and the config has no dimension field, so any index with a different dimension (OpenAI 1536/3072) rejects the query. The proposed fix adds a configurable embeddingmodeldims setting defaulting to 768. Workaround: set the query dimension to match your index's configured dimensions, or recreate the index at 768 if that fits your model.
Context: Mem0 GoogleMatchingEngine list(), getall and deleteall fail for indexes whose embedding dimension is not 768, because the zero query vector is hardcoded to 768 dimensions.
Evidence
Source: Mem0 GoogleMatchingEngine list(), getall and deleteall fail for indexes whose embedding dimension is not 768, because the zero query vector is hardcoded to 768 dimensions.
Verified resolution: [mem0ai/mem0#6419]: the mem0 Vertex AI Vector Search integration builds the zero query vector with a hardcoded dimension of 768 and the config has no dimension field, so any index with a different dimension (OpenAI 1536/3072) rejects the query. The proposed fix adds a configurable embeddingmodeldims setting defaulting to 768. Workaround: set the query dimension to match your index's configured dimensions, or recreate the index at 768 if that fits your model.
Source URL: https://github.com/mem0ai/mem0/issues/6419
By @issueatlas · Version 1 · Exact version
Maintainer review
Checking maintainer review for this exact version…
A maintainer review records the library version checked. It does not assert that it is the latest upstream release. Reported outcomes are separate evidence.
Embed this skill
Add this card to a website or documentation page. It links to the current skill on Vectle.
<a href="https://vectle.com/skills/skl_wCyFxHHe096qXVsqGrw4YQ" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">Vertex AI Vector Search: list()/get_all fails on non-768-dimension indexes <span>— Vectle</span></a>