AlibabaQwen3.7 Plus
Qwen3.7 Plus is an Alibaba proprietary reasoning model, released 2 Jun 2026. Its best configuration ranks #150 of 785 on the BenchLeader Index at 58.5 ±6.3, in the upper half. The ± is the point: 296 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (76) and lowest in coding (44). At $0.700 per million tokens blended it is mid-priced. Output speed of 59 tokens per second puts it slower than most, with a first answer in 36.1 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $0.700/M
- $0.400 in · $1.60 out
- Output speed
- 59 tok/s
- measured by Artificial Analysis
- First answer
- 36 s
- first token 1.23 s
- Context
- 1M
- Full answer
- 45 s
- median, reasoning included
- Cost per run
- $0.217
- one full Intelligence Index run
- Released
- 2 Jun 2026
- Overall index59
- Reasoning60
- Coding44
- Agents & tools45
- Maths65
- Knowledge62
- Instruction following76
- Human preference64
- Multimodal63
- Long context61
- Composite58
Versions
Alibaba has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Qwen3.7 Plusthis page | 2 Jun 2026 | 58.5 | #150 |
| Qwen3.6 Plus | 31 Mar 2026 | 58.8 | #144 |
| Qwen3.5 Plusthinking | 16 Feb 2026 | 57.3 | #182 |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | 59 tok/s | 36 s | $0.0006 | Maths 49 · Reasoning 59 |
| not statedbest | 58.5 | #150 | 59 tok/s | 36 s | $0.0006 | Agents & tools 45 · Coding 44 · Composite 58 · Human preference 64 · Instruction following 76 · Knowledge 62 · Long context 61 · Maths 65 · Multimodal 63 · Reasoning 60 |
How its index has moved
8 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| GPQA Diamond | 81.8%#117 | 87.9%#65 | Epoch AI Benchmarking Hub | |
| Arena Hard Promptseffort not stated | – | 1474#79 | Arena (LMArena) | |
| GPQA Diamond (AA)not in indexeffort not stated | – | 90.0%#66 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in indexeffort not stated | – | 35.6%#117 | Artificial Analysis | |
| CritPteffort not stated | – | 9.1%#128 | Artificial Analysis | |
| Mystery Game Puzzlesnot in index | 17.0%#75 | 17.0%#75 | Epoch AI Benchmarking Hub | |
| LMCAnot in indexeffort not stated | – | 37.6%#117 | Epoch AI Benchmarking Hub | |
| DTBenchnot in indexeffort not stated | – | 84.0%#111 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| SciCode | 46.1%#135 | SciCode | |
| FrontierCode | 10.2%#43 | Cognition | |
| Arena Coding | 1501#79 | Arena (LMArena) | |
| SciCode (AA)not in index | 46.1%#140 | Artificial Analysis | |
| Code Migrationnot in index | 12.9%#60 | Vals AI | |
| Vibe Code Bench v1.1not in index | 46.4%#67 | Vals AI | |
| Terminal-Bench 4.0 (AA)not in index | 1.0%#128 | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 61.0%#96 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| OSWorld-Verified 2.0 | 2.8%#14 | OSWorld | |
| Arena Agent | -6.7#43 | Arena (LMArena) | |
| Terminal-Bench Hard | 47.0%#27 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 93.0%#46 | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals) | 52.8%#55 | Vals AI | |
| GDPval-AA v2.1not in index | 13.3%#108 | Artificial Analysis | |
| τ³-Banking (AA) | 17.5%#106 | Artificial Analysis | |
| APEX-Agents (AA)not in index | 22.4%#20 | Artificial Analysis | |
| Finance Agent v2not in index | 38.2%#62 | Vals AI | |
| Harvey's Legal Agent Benchmark v1not in index | 0.0%#61 | Vals AI | |
| Legal Research Benchnot in index | 16.4%#60 | Vals AI | |
| SkillsBenchnot in index | 54.3%#14 | Vals AI | |
| Tax Agent Benchnot in index | 10.5%#57 | Vals AI | |
| AutomationBenchnot in index | 17.4%#141 | Artificial Analysis | |
| GDP.pdf (AA)not in index | 12.2%#114 | Artificial Analysis | |
| EnterpriseOps-Gymnot in index | 40.6%#27 | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 34.4%#68 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 80.0%#114 | 93.3%#59 | Epoch AI Benchmarking Hub | |
| Arena Mathseffort not stated | – | 1467#56 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-Omniscience | 1.1#126 | Artificial Analysis | |
| Excel Modeling Benchmarknot in index | 49.4%#54 | Vals AI | |
| AA-Omniscience: accuracynot in index | 22.5%#273 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 72.3%#35 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 78.0%#14 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1447#67 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1456#69 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1431#72 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1464#74 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1465#69 | Arena (LMArena) |
Multimodal
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Vision | 1278#37 | Arena (LMArena) | |
| MMMU-Pro | 80.5%#48 | Artificial Analysis | |
| MortgageTaxnot in index | 66.2%#35 | Vals AI | |
| SAGEnot in index | 39.3%#61 | Vals AI |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 73.0%#182 | Artificial Analysis | |
| Arena Documentnot in index | 1453#30 | Arena (LMArena) | |
| MLCRnot in index | 8.9%#122 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 147#69 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 25.2#164 | Artificial Analysis | |
| Vals Multimodal Indexnot in index | 53.9%#22 | Vals AI |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Alibaba Cloud Int. | 47 tok/s | 1.33 s | $0.320 | $1.28 | 1M | – |
| Alibaba Cloud Int. | 43 tok/s | 1.74 s | $0.320 | $1.28 | 1M | – |
Price history
The price above, per 1M tokens, every day it changed.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.080 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0006 | $0.0005 | 41.1 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0058 | $0.0029 | 46.2 s |
| Code edit | 6,000 / 1,500 | $0.0048 | $0.0034 | 1.0 min |
| Agentic coding session | 60,000 / 4,000 | $0.030 | $0.016 | 1.7 min |
| Structured extraction | 2,000 / 200 | $0.0011 | $0.0006 | 39.4 s |
See also
Questions people ask
- What are Qwen3.7 Plus's benchmark scores?
- As of 24 Oct 2026, Qwen3.7 Plus scores 58.5 ±6.3 on the BenchLeader Index, strongest in instruction following (76) and weakest in coding (44). That rests on 20 independent benchmark results, and every score on this page links to its publisher.
- How does Qwen3.7 Plus rank?
- Qwen3.7 Plus is #150 of 785 ranked configurations on 24 Oct 2026. 296 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Qwen3.7 Plus cost per 1M tokens?
- As of 24 Oct 2026, Qwen3.7 Plus costs $0.400 per 1M input tokens and $1.60 per 1M output tokens, $0.700 blended (three input tokens per output token). The list price last changed on 28 Sept 2026.
- When was Qwen3.7 Plus released?
- Alibaba released Qwen3.7 Plus on 2 Jun 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Qwen3.7 Plus: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/qwen3-7-plus.