Qwen3.7 Plus vs Qwen3.8 Flash
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Qwen3.7 Plus35.6
Qwen Flash38
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Qwen3.7 Plus1
Qwen Flash25.3
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Qwen3.7 Plus17.4
Qwen Flash55.9
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Qwen3.7 Plus1,431
Qwen Flash—
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos
Qwen3.7 Plus80.5
Qwen Flash79.8
| Benchmark | Qwen3.7 Plus | Qwen Flash |
|---|---|---|
| Humanity's Last Exam | 35.6 | 38 |
| Terminal-Bench 4.0 | 1 | 25.3 |
| AutomationBench-AA | 17.4 | 55.9 |
| Arena Text · creative writing | 1,431 | — |
| MMMU-Pro (vision) | 80.5 | 79.8 |
Specs
| Spec | Qwen3.7 Plus | Qwen Flash |
|---|---|---|
| Input, per 1M | $0.32 | $0.15 |
| Output, per 1M | $1.28 | $0.47 |
| Cost per task? | $0.22 | $0.37 |
| Speed, tok/s | 58 | 55 |
| Context | 1M | 262K |
| Weights | Closed | Open |
| Released | Jun 2026 | Aug 2026 |
| Memory (4-bit) | — | 108 GB |
| Licence | — | qwen-community-1.0 |
| Commercial use | — | Restricted |
| See images | ✓ | ✓ |
| Read PDFs | Partial | Partial |
| Hear audio | — | — |
| Watch video | — | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Qwen3.7 Plus → Qwen Flash
Context: 1M → 262K
max_tokens range · QwenCloud›
With thinking enabled, max_tokens outside 1–32768 returns HTTP 400; use max_completion_tokens to budget reasoning plus answer without that cap.
File is not delivered · QwenCloud / Responses›
PDF input through Responses returns HTTP 200 without delivering the file to the model; use Chat Completions or DashScope.
Smaller values automatically adjusted · QwenCloud›
In Flash thinking mode, temperature below 0.6 is automatically raised to 0.6.
Static YaRN affects shorter texts · Self-hosted›
Static YaRN can hurt short-text performance; change rope_parameters only when longer contexts are required.
tool_choice only supports auto or none · QwenCloud›
With thinking enabled, tool_choice supports only auto or none.
Include the word JSON · QwenCloud›
JSON Object mode requires the word JSON in a system or user message; otherwise the API returns an error.
Qwen Flash → Qwen3.7 Plus
Nothing known.