# CoIR · Code Information Retrieval

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/mteb%3ACoIR. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/mteb%3ACoIR.

Code information retrieval across diverse programming languages and coding tasks, including code search, question answering, and text-to-SQL retrieval.

Canonical URL: https://benchize.com/benchmarks/mteb%3ACoIR
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Code information retrieval across diverse programming languages and coding tasks, including code search, question answering, and text-to-SQL retrieval.

Embeddings · Models tested: 2 · Score publishers: [MTEB](https://github.com/CoIR-team/coir)

## Model results

| # | Model | Score |  |
| --- | --- | --- | --- |
| 1 | [Qwen3 Embedding](https://benchize.com/models/qwen3-embedding.md) | 0.8source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.79864 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.79864 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 8 scoresByTask {"AppsRetrieval":0.91071,"COIRCodeSearchNetRetrieval":0.8951116666666667,"CodeFeedbackMT":0.93699,"CodeFeedbackST":0.89929,"CodeSearchNetCCRetrieval":0.9634683333333333,"CodeTransOceanContest":0.93733,"CodeTransOceanDL":0.32809,"CosQA":0.38042,"StackOverflowQA":0.94751,"SyntheticText2SQL":0.78748} scoresByTaskType {"Retrieval":0.79864} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/CoIR/scores) Source MTEB Original model label Qwen/Qwen3-Embedding-8B Observed 2026-10-21T16:40:38.655Z Evidence SHA-256 `702b3cd3c4c0e45aa8a44651b56cf0945a7291e9b615807d81f74c44ee69a8bf` Source locator `/rows/7` |  |  |  |
| 2 | [Qwen3 Embedding](https://benchize.com/models/qwen3-embedding.md) | 0.79source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.7919594999999999 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.7919594999999999 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 7 scoresByTask {"AppsRetrieval":0.89176,"COIRCodeSearchNetRetrieval":0.8792833333333334,"CodeFeedbackMT":0.93211,"CodeFeedbackST":0.89514,"CodeSearchNetCCRetrieval":0.9558516666666667,"CodeTransOceanContest":0.90993,"CodeTransOceanDL":0.35039,"CosQA":0.37982,"StackOverflowQA":0.94323,"SyntheticText2SQL":0.78208} scoresByTaskType {"Retrieval":0.7919594999999999} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/CoIR/scores) Source MTEB Original model label Qwen/Qwen3-Embedding-4B Observed 2026-10-21T16:40:38.655Z Evidence SHA-256 `702b3cd3c4c0e45aa8a44651b56cf0945a7291e9b615807d81f74c44ee69a8bf` Source locator `/rows/6` |  |  |  |
| 3 | [Gemini Embedding 001](https://benchize.com/models/gemini-embedding-001.md) | 0.74source score |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `mean_task` \| 0.7390116666666666 \| source_score \| source_published_aggregate \| Higher is better \| \| `mean_task_type` \| 0.7390116666666666 \| source_score \| source_published_aggregate \| Higher is better \| ### Configuration zeroShotPct 100 ### Reported statistics experiments Not reported rank 18 scoresByTask {"AppsRetrieval":0.93751,"COIRCodeSearchNetRetrieval":0.8105950000000001,"CodeFeedbackMT":0.56279,"CodeFeedbackST":0.85329,"CodeSearchNetCCRetrieval":0.8468616666666667,"CodeTransOceanContest":0.89529,"CodeTransOceanDL":0.31467,"CosQA":0.50238,"StackOverflowQA":0.96713,"SyntheticText2SQL":0.6996} scoresByTaskType {"Retrieval":0.7390116666666666} trainedOnTasks Not reported ### Source [Original source](https://mteb-leaderboard-backend.hf.space/v1/benchmarks/CoIR/scores) Source MTEB Original model label google/gemini-embedding-001 Observed 2026-10-21T16:40:38.655Z Evidence SHA-256 `702b3cd3c4c0e45aa8a44651b56cf0945a7291e9b615807d81f74c44ee69a8bf` Source locator `/rows/17` |  |  |  |

---
Cite as: CoIR · Code Information Retrieval — Benchize, https://benchize.com/benchmarks/mteb%3ACoIR (retrieved 2026-10-21)