Qwen 3.7 Max
Language modelAPI only
Qwen 3.7 Max
Alibaba
- Released
- May 2026
- Data date
- September 8, 2026
- Speed
- 199.8 tokens/sOutput throughput
- Price
- Input · Output / 1M tokens
- Input
- Text
- Output
- Text
Qwen 3.7 Max combines a 1,000,000-token window with Alibaba's focus on coding, office automation, MCP integrations, and long autonomous runs. Its integrated thinking mode can be controlled through the API. The parameter count remains undisclosed, and the weights are not downloadable.
Alibaba’s current documentation places qwen3.7-max in legacy tables, while still documenting the endpoint and dated snapshots. Qwen 3.8 Max belongs to a separately documented Qwen3.8 serving line.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | ProprietarySource |
| Access | APISource |
| Context window | 1,000,000 tokensSource |
| Notes | Qwen flagship of the 3.7 generation, released May 21, 2026 and superseded by Qwen3.8-Max on September 2, 2026. 1M context (served ~991K), integrated thinking mode, API-only via Alibaba Cloud Model Studio (closed-weight). Parameter count not disclosed.Source |
| API model ID | qwen3.7-maxSource |
| Input | TextSource |
| Output | TextSource |
| Reasoning mode | reasoningSource |
| Tool calling | YesSource |
| Structured output | YesSource |
| Maximum context window | 1,000,000 tokensSource |
| Release | 2026-05-21Source |
Page 1 of 17
Published benchmarks
Cybersecurity
SecCodeBench-V2
98 executable project cases across 22 CWE categories test functionality before security. The weighted score combines generation, repair, native prompts, and security-aware prompts.
| Sub-task | Qwen 3.7 Max (This model) | Claude Opus 4.5 | Claude Opus 4.6 | GPT-5.4 |
|---|---|---|---|---|
| Repair without hints | 72.5% | 73.67% (best value in this row) | 73.1% | 61.41% |
| Repair with security hints | 86.77% (best value in this row) | 82.18% | 81.73% | 80.04% |
| Generation without hints | 55.12% | 58.02% (best value in this row) | 51.05% | 50.97% |
| Generation with security hints | 70.85% | 72.1% (best value in this row) | 70.65% | 67.82% |
| Weighted total | 66.81% | 68.1% (best value in this row) | 64.9% | 59.74% |
higher is betterAs of August 9, 2026Source: SecCodeBenchbars scaled within each row, not across rows
Coding
Code Migration
| Sub-task | Qwen 3.7 Max (This model) | Qwen 3.7 Plus |
|---|---|---|
| CLI | 11.63% | 11.86% (best value in this row) |
| COBOL | 17.4% (best value in this row) | 15.84% |
| Overall | 13.07% (best value in this row) | 12.86% |
higher is betterAs of October 15, 2026Source: Vals AIbars scaled within each row, not across rows
Methodology and context
- Metric
- accuracy
Code Migration
Code Quality
higher is betterAs of October 15, 2026Scale: 0 rating - 3.44 rating
Methodology and context
- Comparison cohort
- vals-code-migration:1:code_quality
- Metric
- mean reviewer code-quality rating
- Scale
- No upper bound is documented. The chart spans the recorded values and zero.
- Participants
- 2
- Methodology and context
- Source model ID: alibaba/qwen3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-12. Temperature: 0.7. Maximum output tokens: 65536. Latency in seconds: 1531.852. Standard error: 0.045. Cost per task in USD: 1.932899. Score display: 3.437 / 5. Source ID: vals-ai. Evidence capture ID: vals-code-migration-2026-10-15-3fa9c10568ec. Source payload hash: 3fa9c10568ecc117fabec81767c405b49a25d5d2ded308ee435fffb197c53474. Parser version: vals-astro-v3. Imported at: 2026-10-15
Recorded measurement details
SecCodeBench-V2: Repair without hints (Show measurement, test conditions, and source)
- Source value
- 72.5
- Score
- 72.5
- Metric
- native repair pass@1
- Unit
- %
- Benchmark version
- 2.2.0
- Category
- cybersecurity
- Task
- fix
- Task label
- Repair without hints
- Direction
- Higher is better
- Comparison cohort
- seccodebench-v2-2-0:official:fix
- Source type
- official-leaderboard
- Evaluator
- SecCodeBench
- Status
- active
- Retrieved at
- 2026-08-09
- Methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.. Source model ID: curated/qwen-3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: SecCodeBench. Dataset type: public. Leaderboard date: 2026-02-16. Reasoning effort: thinking. Source ID: seccodebench. Evidence capture ID: curated-seccodebench-2026-08-09-695eb9d33cf9. Source payload hash: 695eb9d33cf912caeb0e55f962a56043ed10df54d22fefdf727d2d0e4fdc407d. Parser version: curated-v1. Imported at: 2026-10-15
- Comparison methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.
SecCodeBench-V2: Repair with security hints (Show measurement, test conditions, and source)
- Source value
- 86.77
- Score
- 86.77
- Metric
- security-aware repair pass@1
- Unit
- %
- Benchmark version
- 2.2.0
- Category
- cybersecurity
- Task
- fixHints
- Task label
- Repair with security hints
- Direction
- Higher is better
- Comparison cohort
- seccodebench-v2-2-0:official:fixHints
- Source type
- official-leaderboard
- Evaluator
- SecCodeBench
- Status
- active
- Retrieved at
- 2026-08-09
- Methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.. Source model ID: curated/qwen-3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: SecCodeBench. Dataset type: public. Leaderboard date: 2026-02-16. Reasoning effort: thinking. Source ID: seccodebench. Evidence capture ID: curated-seccodebench-2026-08-09-7b4c8460fb2a. Source payload hash: 7b4c8460fb2a0ba876a91bbe99cb951a921caffc70c2f305c0dc44aab174d352. Parser version: curated-v1. Imported at: 2026-10-15
- Comparison methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.
SecCodeBench-V2: Generation without hints (Show measurement, test conditions, and source)
- Source value
- 55.12
- Score
- 55.12
- Metric
- native generation pass@1
- Unit
- %
- Benchmark version
- 2.2.0
- Category
- cybersecurity
- Task
- generation
- Task label
- Generation without hints
- Direction
- Higher is better
- Comparison cohort
- seccodebench-v2-2-0:official:generation
- Source type
- official-leaderboard
- Evaluator
- SecCodeBench
- Status
- active
- Retrieved at
- 2026-08-09
- Methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.. Source model ID: curated/qwen-3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: SecCodeBench. Dataset type: public. Leaderboard date: 2026-02-16. Reasoning effort: thinking. Source ID: seccodebench. Evidence capture ID: curated-seccodebench-2026-08-09-f686ae9d8174. Source payload hash: f686ae9d81746759402d2ad46ea834a7a66a5a2a14674a3c903d13278a745fe2. Parser version: curated-v1. Imported at: 2026-10-15
- Comparison methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.
SecCodeBench-V2: Generation with security hints (Show measurement, test conditions, and source)
- Source value
- 70.85
- Score
- 70.85
- Metric
- security-aware generation pass@1
- Unit
- %
- Benchmark version
- 2.2.0
- Category
- cybersecurity
- Task
- generationHints
- Task label
- Generation with security hints
- Direction
- Higher is better
- Comparison cohort
- seccodebench-v2-2-0:official:generationHints
- Source type
- official-leaderboard
- Evaluator
- SecCodeBench
- Status
- active
- Retrieved at
- 2026-08-09
- Methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.. Source model ID: curated/qwen-3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: SecCodeBench. Dataset type: public. Leaderboard date: 2026-02-16. Reasoning effort: thinking. Source ID: seccodebench. Evidence capture ID: curated-seccodebench-2026-08-09-a222a11cf867. Source payload hash: a222a11cf867d891c319011878debf544c874e4fc3dd30533bb3ede2492dd03b. Parser version: curated-v1. Imported at: 2026-10-15
- Comparison methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.
SecCodeBench-V2: Weighted total (Show measurement, test conditions, and source)
- Source value
- 66.81
- Score
- 66.81
- Metric
- weighted pass@1
- Unit
- %
- Benchmark version
- 2.2.0
- Category
- cybersecurity
- Task
- total
- Task label
- Weighted total
- Direction
- Higher is better
- Comparison cohort
- seccodebench-v2-2-0:official:total
- Source type
- official-leaderboard
- Evaluator
- SecCodeBench
- Status
- active
- Retrieved at
- 2026-08-09
- Methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.. Source model ID: curated/qwen-3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: SecCodeBench. Dataset type: public. Leaderboard date: 2026-02-16. Reasoning effort: thinking. Source ID: seccodebench. Evidence capture ID: curated-seccodebench-2026-08-09-08df712fc024. Source payload hash: 08df712fc02407c43cec90b69803b9fcd219c0a9436460f0ffa5fecc003d9284. Parser version: curated-v1. Imported at: 2026-10-15
- Comparison methodology
- 98 project cases. Functional tests run before dynamic exploit checks and security scoring.
Code Migration: CLI (Show measurement, test conditions, and source)
- Source value
- 11.625
- Score
- 11.625
- Metric
- accuracy
- Unit
- %
- Benchmark version
- 1
- Category
- coding
- Task
- cli
- Task label
- CLI
- Direction
- Higher is better
- Comparison cohort
- vals-code-migration:1:cli
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Source model ID: alibaba/qwen3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-12. Temperature: 0.7. Maximum output tokens: 65536. Latency in seconds: 1613.281. Standard error: 1.536. Cost per task in USD: 1.981447. Source ID: vals-ai. Evidence capture ID: vals-code-migration-2026-10-15-3fa9c10568ec. Source payload hash: 3fa9c10568ecc117fabec81767c405b49a25d5d2ded308ee435fffb197c53474. Parser version: vals-astro-v3. Imported at: 2026-10-15
- Comparison methodology
- Temperature: 0.7. Maximum output tokens: 65536
Code Migration: COBOL (Show measurement, test conditions, and source)
- Source value
- 17.403
- Score
- 17.403
- Metric
- accuracy
- Unit
- %
- Benchmark version
- 1
- Category
- coding
- Task
- cobol
- Task label
- COBOL
- Direction
- Higher is better
- Comparison cohort
- vals-code-migration:1:cobol
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Source model ID: alibaba/qwen3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-12. Temperature: 0.7. Maximum output tokens: 65536. Latency in seconds: 1287.566. Standard error: 10.464. Cost per task in USD: 1.787255. Source ID: vals-ai. Evidence capture ID: vals-code-migration-2026-10-15-3fa9c10568ec. Source payload hash: 3fa9c10568ecc117fabec81767c405b49a25d5d2ded308ee435fffb197c53474. Parser version: vals-astro-v3. Imported at: 2026-10-15
- Comparison methodology
- Temperature: 0.7. Maximum output tokens: 65536
Code Migration: Overall (Show measurement, test conditions, and source)
- Source value
- 13.069
- Score
- 13.069
- Metric
- accuracy
- Unit
- %
- Benchmark version
- 1
- Category
- coding
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Comparison cohort
- vals-code-migration:1:overall
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Source model ID: alibaba/qwen3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-12. Temperature: 0.7. Maximum output tokens: 65536. Latency in seconds: 1531.852. Standard error: 2.858. Cost per task in USD: 1.932899. Source ID: vals-ai. Evidence capture ID: vals-code-migration-2026-10-15-3fa9c10568ec. Source payload hash: 3fa9c10568ecc117fabec81767c405b49a25d5d2ded308ee435fffb197c53474. Parser version: vals-astro-v3. Imported at: 2026-10-15
- Comparison methodology
- Temperature: 0.7. Maximum output tokens: 65536
Code Migration: Code Quality (Show measurement, test conditions, and source)
- Source value
- 3.437
- Score
- 3.437
- Metric
- mean reviewer code-quality rating
- Unit
- rating
- Benchmark version
- 1
- Category
- coding
- Task
- code_quality
- Task label
- Code Quality
- Direction
- Higher is better
- Comparison cohort
- vals-code-migration:1:code_quality
- Source type
- independent-evaluator
- Evaluator
- Vals AI
- Status
- active
- Retrieved at
- 2026-10-15
- Methodology
- Source model ID: alibaba/qwen3.7-max. Source model name: Qwen 3.7 Max. Provider reported by source: Alibaba. Evaluator: Vals AI. Dataset type: private. Leaderboard date: 2026-10-12. Temperature: 0.7. Maximum output tokens: 65536. Latency in seconds: 1531.852. Standard error: 0.045. Cost per task in USD: 1.932899. Score display: 3.437 / 5. Source ID: vals-ai. Evidence capture ID: vals-code-migration-2026-10-15-3fa9c10568ec. Source payload hash: 3fa9c10568ecc117fabec81767c405b49a25d5d2ded308ee435fffb197c53474. Parser version: vals-astro-v3. Imported at: 2026-10-15
- Comparison methodology
- Temperature: 0.7. Maximum output tokens: 65536
Page 1 of 17
Our tests
SVG short test
One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.
Earlier tests
Pelican on a bicycle
A community classic for free-form SVG drawing.
Qwen 3.7 Max
Reasoning (requested): high
Current attempt
- Run date
- Oct 1, 2026
- API provider
- openrouter
- Requested API model
- qwen/qwen3.7-max
- API model reported in response
- qwen/qwen3.7-max
- Reasoning setting
- high
- Pinned endpoint
- Alibaba | qwen/qwen3.7-max-20260520
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Passed, not a quality rating
- Response time
- 28 s
Visually reviewed. This individual test does not establish overall model quality.
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)Popularity
45.1 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Speed | Alibaba. Output speed after the first streaming chunk, not total response time. Classified as deprecated by Artificial Analysis. Artificial Analysis maintains only the default 10K-input workload there. Retrieved September 8, 2026. |
| Price | Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. |
| Input | Documented modalities. Retrieved September 8, 2026. |
| Output | Documented modalities. Retrieved September 8, 2026. |
| Research status | Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | knowledge-cutoff Original wording Checked Model Studio model/API pages do not state an exact knowledge cutoff. |
| Not found in the inspected sources | profile-benchmarks Original wording No exact-model numeric benchmark table was found in the checked API documentation. |
| Not found in the inspected sources | parameters-billions Original wording The checked API documentation does not publish total parameters. |
| Unresolved | input-modalities Original wording The checked API model list confirms text generation but the vision table does not provide a qwen3.7-max row; no multimodal claim made. |
| Not published | benchmark run metadata Note Qwen's report does not give evaluation task dates or reasoning effort, and most rows omit metric units and row-level setup. The configuration retains each disclosed row-specific detail without inferring omitted values. |
| Additional source | Model metadata · Alibaba Qwen3.7-Max announcement · Editorial description reviewed October 2, 2026 |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | Artificial Analysis (retrieved October 15, 2026) |
| Additional source | SecCodeBench (retrieved August 9, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Vals AI (retrieved October 15, 2026) |
| Additional source | Qwen Team Qwen3.7 report (retrieved October 15, 2026) |
| Additional source | Model metadata (retrieved September 8, 2026) |
| Additional source | Alibaba Cloud text generation models · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |