mxbai-rerank-v3.1-listwise
SearchOpen menu
On this page<br>Excellent and fasterTry it now
[data-slot=icon-swap]]:px-3 group mb-10 h-4 self-start p-0 hover:bg-transparent has-[>svg]:px-0" href="/blog">All postsmxbai-rerank-v3.1-listwise
svg]:pointer-events-none [&>svg]:size-3 border-transparent bg-secondary text-secondary-foreground [a&]:hover:bg-secondary/90">Researchsvg]:pointer-events-none [&>svg]:size-3 border-transparent bg-secondary text-secondary-foreground [a&]:hover:bg-secondary/90">Product<br>MB<br>Mixedbread Team<br>·July 27, 2026·4 min read
Copy to clipboard
mxbai-rerank-v3.1-listwise is now available as the default reranker.
It delivers gpt-5.6-sol-level ranking quality with substantially improved latency over mxbai-rerank-v3-listwise. It is excellent at complex tasks like recency-aware ranking, source-priority resolution, and multi-step composite instructions, compared with leading LLMs and pointwise rerankers.
Quality versus latency on ViDoRe v3: mxbai-rerank-v3.1-listwise reaches gpt-5.6-sol quality at roughly 61 times lower latency, while Cohere and the GPT-5.6 rerankers are slower.v3.1 reaches gpt-5.6-sol (high) quality while being 61× faster
0.600.640.680.72100s10s1s0.1sNDCG@10Latency per query (log scale, lower is faster)First stage+ cohere rerank-4-pro+ gpt-5.6-terra (high)+ gpt-5.6-sol (high)+ gpt-5.6-luna (high)+ mxbai-rerank-v3.1-listwise
Sampled queries per ViDoRe subset with Mixedbread base search (Wholembed v3) as first stage.
Excellent and fasterLink to section
Unlike pointwise rerankers measuring the relevance of individual documents, our listwise v3 reads the whole candidate set for better reasoning at the cost of higher ranking latency.
Now, by rewriting its underlying inference engine, v3.1 reranks faster across every input size, matching pointwise rerankers. In production, the speedup ranges from ~25% on typical queries to ~54% on long-tail inputs (64–128k tokens).
Median reranking latency by input size: mxbai-rerank-v3.1-listwise is faster than v3 at every input size.v3.1 ranks 25–54% faster than v3 across input sizes
02004006008–16k16–32k32–64kMedian latency (ms)Input tokensv3v3.1
For better threshold cutoffs, we also changed v3.1 to return content-dependent relevance scores instead of a fixed rank-based ladder in v3.
Try it nowLink to section
mxbai-rerank-v3.1-listwise is available today through Mixedbread Search:
PythonTypeScript<br>figure:only-child]:mt-0" style="animation-duration:0s">Copy to clipboardpre]:bg-inherit [&>code]:rounded-none [&>code]:border-0 [&>code]:bg-transparent [&>code]:p-0 [&_.line]:pl-0! shiki shiki-themes github-light github-dark" style="--shiki-light:#24292e;--shiki-dark:#e1e4e8;--shiki-light-bg:#fff;--shiki-dark-bg:#24292e" tabindex="0">from mixedbread import Mixedbread
client = Mixedbread()
results = client.stores.search(<br>store_identifiers=["my-store"],<br>query="when is my flight to London? The most recent valid booking wins",<br>search_options={<br>"rerank": {<br>"model": "mixedbread-ai/mxbai-rerank-v3.1-listwise",<br>},
figure:only-child]:mt-0">
[data-slot=icon-swap]]:px-3 has-[>svg]:px-3 mr-auto flex w-auto items-center justify-center sm:justify-start" href="/blog/infinite-file-sizes">Building a Robust Ingestion Sy...
Toggle theme