qwen
Qwen3 Reranker 0.6B
qwen/qwen3-reranker-0.6b
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Qwen3 Reranker 0.6B is the smallest text reranking model in Alibaba Cloud's Qwen3 Reranker series, built on the Qwen3 architecture and derived from the Qwen3-0.6B base. It scores query-document pairs for relevance rather than generating text, with a 32K context length and reported support for more than 100 languages, including programming languages. The model accepts task-specific instructions, which the publisher says typically improve results. Released under Apache 2.0, it works with Sentence Transformers, Transformers, and vLLM.[1][2]
Strengths
- Publisher documentation reports support for over 100 languages, including programming languages, enabling multilingual, cross-lingual, and code retrieval scenarios.[2]
- Instruction-aware design lets developers customize task instructions; the publisher reports this typically yields a modest improvement over uninstructed use.[2]
- Integrates with standard tooling such as Sentence Transformers, Transformers (>=4.51.0), and vLLM, with a 32K context window suited to long documents.[1][2]
Limitations
- Vendor-published results show the 0.6B variant trailing its larger 4B and 8B siblings on most reported reranking metrics; no independent third-party evaluation was included in this source pack.[2]
- Practical use requires recent library versions (transformers>=4.51.0, vllm>=0.8.5), and the publisher notes instructions tend to work best when written in English.[2]
- Evidence is limited to the vendor's model card and an OpenRouter catalog listing; ranking quality and latency on your own data need workload validation.[1][2]
Where to use it
- Second-stage reranking to re-order candidate lists from a first-pass retriever, for example paired with a Qwen3 embedding model as in the vendor's own evaluation setup.[2]
- Multilingual or code search deployments where a compact, instruction-tunable reranker is preferred over a larger model.[2]
- Latency- or resource-sensitive pipelines where the smallest model in the series is a reasonable starting point, subject to on-task validation.[1][2]
Sources
- [1] OpenRouter catalog · Qwen3 Reranker 0.6BSources checked: Nov 13, 2026
- [2] Model documentation · Qwen3 Reranker 0.6BSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Qwen3 Reranker 0.6B is the smallest text reranking model from Alibaba Cloud in the Qwen3 Reranker series. Built on the Qwen3 architecture, it evaluates query-document pairs to produce relevance scores...
Specifications
- Model ID
- qwen/qwen3-reranker-0.6b
- Creator
- qwen
- Added to OpenRouter
- 26.08.2026
- Context length
- 32,768
- Modality
- text->rerank
- Input modalities
- text
- Output modalities
- rerank
- Token pricing
- No per-token price in the source snapshot — see OpenRouter for the billing unit
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | — |
| Coding | No measurement | — |
| Agents | No measurement | — |
| Writing | No measurement | — |
| Reasoning | No measurement | — |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| PrimeIntellectbf16 | 32,768 | — | — | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.