MTEB(eng, v2): MTEB English v2
English text-embedding benchmark from the Massive Text Embedding Benchmark project. Covers 41 tasks across retrieval, reranking, classification, clustering, pair classification, semantic similarity and summarization, chosen to limit overlap with common fine-tuning data. The score is the mean of per-task main scores. Higher is better.
Models ranked
11
tracked in MTEB(eng, v2)
Score band
75.2 – 62.8
best → lowest tracked
Snapshot trend
—
MTEB(eng, v2) · need ≥2 snapshots
Published snapshot
MTEB(eng, v2) · observed Oct 17, 2026 · 11 model variants.
Notes: MTEB(eng, v2) Mean (Task) 75.22 as published on the Qwen3-Embedding model card (standing rule: the model's own card wins). The leaderboard aggregation of embeddings-benchmark/results @ 4bb41dd gives 75.23 on 2026-10-17 (rounding only). Source: Qwen3-Embedding card; MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17.
Notes: MTEB(eng, v2) Mean (Task) 73.30 over 41 tasks; Mean (TaskType) 67.67. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17.
Notes: MTEB(eng, v2) Mean (Task) 70.70 as published on the Qwen3-Embedding model card (standing rule: the model's own card wins). The leaderboard aggregation of embeddings-benchmark/results @ 4bb41dd gives 70.47 on 2026-10-17; the difference is larger than rounding, likely re-run results added to the repo after the card was written. Source: Qwen3-Embedding card; MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17.
Notes: MTEB(eng, v2) Mean (Task) 69.53 over 41 tasks; Mean (TaskType) 64.82. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17.
Notes: MTEB(eng, v2) Mean (Task) 66.21 over 41 tasks; Mean (TaskType) 61.77. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 66.01 over 41 tasks; Mean (TaskType) 61.43. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 65.89 over 41 tasks; Mean (TaskType) 61.87. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 65.14 over 41 tasks; Mean (TaskType) 60.77. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 64.59 over 41 tasks; Mean (TaskType) 60.57. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 64.30 over 41 tasks; Mean (TaskType) 59.93. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Notes: MTEB(eng, v2) Mean (Task) 62.75 over 41 tasks; Mean (TaskType) 58.49. Computed with the leaderboard's own code (mteb 2.26.6 Benchmark summary table) from embeddings-benchmark/results @ 4bb41dd (2026-10-17). Source: MTEB (embeddings-benchmark/results, CC0). Retrieved 2026-10-17. Results folder includes an 'external' (copied/reported) revision alongside maintainer-run results.
Attribution. Source: MTEB (embeddings-benchmark/results, CC0)
Multilingual edition: Enevoldsen et al., MMTEB (arXiv:2502.13595).
How to read this benchmark
This benchmark scores models where higher is better. Use scores for directional filtering and shortlisting, not universal quality ranking; then validate pricing, context window, provider availability, and fit for your workload.
Trust this score when
- There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
- The model list covers the same version family you can actually deploy today.
- Top candidates overlap with your required routing and feature requirements.
Be cautious when
- There is only one benchmark snapshot or the dataset appears stale.
- The benchmark metric direction is opposite of your decision objective.
- The score difference between options is narrow and likely within implementation variance.
Related benchmarks
Last reviewed: Oct 17, 2026