Model Details
Qwen3.7 Plus
Quick answer: Qwen3.7 Plus is Alibaba's March 2026 mid-tier API model scoring 88.5% MMLU-Pro, 77.7% SWE-Bench Verified, and 34.7% HLE. Below Qwen3.7 Max (89.6% MMLU-Pro) but more cost-efficient.
At a Glance
Where Qwen3.7 Plus leads
- 88.5% MMLU-Pro — near-equal to Max (89.6%)
- 77.7% SWE-Bench Verified — near-equal to Max (80.4%)
- Lower cost than Qwen3.7 Max
Where it lags
- 34.7% HLE — below Max (41.4%)
- Proprietary API only
Best for: Cost-optimised Qwen3.7 generation deployments; MMLU-Pro/SWE-Bench Verified tasks where Max is over-budget.
What Qwen3.7 Plus Is
Qwen3.7 Plus is the "Plus" (mid) tier of Alibaba's March 2026 Qwen3.7 API generation, below the "Max" tier. It sacrifices ~5% HLE for lower cost while maintaining near-equal MMLU-Pro (88.5% vs 89.6%) and SWE-Bench (77.7% vs 80.4%).
Specifications
| Field | Value |
|---|---|
| Organization | Alibaba |
| License | Proprietary (API only) |
| Release date | March 2026 |
| Modality | Text only |
Pricing
Available via Alibaba Cloud Qwen API at lower cost than Qwen3.7 Max.
Public Benchmark Scores
| Benchmark | Score | Source | Date |
|---|---|---|---|
| MMLU-Pro | 88.5% | Benchgen evaluation | 2026-03 |
| SWE-Bench Verified | 77.7% | Benchgen evaluation | 2026-03 |
| Humanity's Last Exam | 34.7% | Benchgen evaluation | 2026-03 |
Qwen3.7 Plus vs Alternatives
| Model | MMLU-Pro | SWE-Bench Verified | HLE | License |
|---|---|---|---|---|
| Qwen3.7 Plus | 88.5% | 77.7% | 34.7% | Proprietary |
| Qwen3.7 Max | 89.6% | 80.4% | 41.4% | Proprietary |
| Qwen3.6 27B | 86.2% | 77.2% | — | Apache 2.0 |
For open-source equivalent: Qwen3.6 27B (Apache 2.0) is close on both MMLU-Pro (86.2%) and SWE-Bench (77.2%).
Related Models & Benchmarks
- Compare: Qwen3.7 Max · Qwen3.6 27B
- Benchmarks: MMLU-Pro · SWE-Bench Verified
Frequently Asked Questions
What is Qwen3.7 Plus?
Alibaba's March 2026 mid-tier API model scoring 88.5% MMLU-Pro, 77.7% SWE-Bench Verified. Proprietary — cost-efficient alternative to Qwen3.7 Max.Specs from Alibaba's Qwen3.7 Plus release (March 2026) and Benchgen evaluations. Last updated 2026-07-24.
Benchmark Leaderboards
This model isn’t on any benchmark leaderboard yet.