# GPT-5.2 vs Qwen3-VL-235B-A22B Instruct

> Side-by-side benchmark comparison for GPT-5.2 and Qwen3-VL-235B-A22B Instruct across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/gpt-5-2-vs-qwen3-vl-235b-a22b-instruct
- Last updated: November 18, 2026

- Shared sourced benchmarks: 6
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Pick Qwen3-VL-235B-A22B Instruct if you want the stronger benchmark profile. GPT-5.2 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Qwen3-VL-235B-A22B Instruct leads overall 60 to 57.83.
- The biggest single benchmark swing is CC-OCR in Multimodal & Grounded, with scores of 70.3% and 82.2%.
- GPT-5.2 also has the larger context window at 400K.

## Model Snapshot

| Property | GPT-5.2 | Qwen3-VL-235B-A22B Instruct |
|----------|----------|----------|
| Creator | OpenAI | Alibaba |
| Type | Proprietary | Open Weight |
| Reasoning | Reasoning | Non-Reasoning |
| Context | 400K | 256K |
| Overall Score | 57.83 | null |
| Benchmarks Covered | 38 | 8 |
| Pricing (input/output) | $1.75 / $14.00 | N/A |

## Category Breakdown

### Agentic

- Winner: Not comparable
- GPT-5.2 public-lane score: 43.5 (Supported · #65/138)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| LifeVal | 49.3% | Coming soon | Coming soon |
| BrowseComp | 65.8% | Coming soon | Coming soon |
| OSWorld-Verified | 47.3% | Coming soon | Coming soon |
| Gert Labs | 46.54% | Coming soon | Coming soon |
| JobBench | 34.3% | Coming soon | Coming soon |

### Coding

- Winner: Not comparable
- GPT-5.2 public-lane score: 37.4 (Supported · #78/160)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80% | Coming soon | Coming soon |
| SWE-bench Pro | 55.6% | Coming soon | Coming soon |
| Vibe Code Bench | 53.50% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 85.4% | Coming soon | Coming soon |
| SWE-bench (Vals) | 75.8% | Coming soon | Coming soon |
| PostTrainBench v1 | 21.4% | Coming soon | Coming soon |

### Reasoning

- Winner: Not comparable
- GPT-5.2 public-lane score: 50 (Supported · #52/159)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| BioBasicsBench | 31.2% | Coming soon | Coming soon |
| ARC-AGI-2 | 52.9% | Coming soon | Coming soon |
| ARC-AGI-1 | 86.17% | Coming soon | Coming soon |

### Multimodal & Grounded

- Winner: Not comparable
- GPT-5.2 public-lane score: 57.8 (Estimated · #40/66)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| MMMU | 86.7% | 78.7% | GPT-5.2 |
| MMMU-Pro | 79.5% | 68.1% | GPT-5.2 |
| RealWorldQA | 83.3% | 79.3% | GPT-5.2 |
| OmniDocBench 1.5 | 85.7% | Coming soon | Coming soon |
| Video-MME (with subtitle) | 86.0% | Coming soon | Coming soon |
| Video-MME (w/o subtitle) | 85.8% | 79.2% | GPT-5.2 |
| MathVision | 83.0% | Coming soon | Coming soon |
| We-Math | 79.0% | Coming soon | Coming soon |
| DynaMath | 86.8% | Coming soon | Coming soon |
| SimpleVQA | 55.8% | Coming soon | Coming soon |
| CC-OCR | 70.3% | 82.2% | Qwen3-VL-235B-A22B Instruct |
| AI2D_TEST | 92.2% | Coming soon | Coming soon |
| CountBench | 91.9% | Coming soon | Coming soon |
| ERQA | 59.8% | Coming soon | Coming soon |
| VideoMMMU | 85.9% | 74.7% | GPT-5.2 |
| MLVU (M-Avg) | 85.6% | Coming soon | Coming soon |
| CharXiv | 82.1% | Coming soon | Coming soon |
| V* | 75.9% | Coming soon | Coming soon |
| OCRBench V2 | Coming soon | 67.1% | Coming soon |
| ScreenSpot Pro | Coming soon | 62.0% | Coming soon |

### Knowledge

- Winner: Not comparable
- GPT-5.2 public-lane score: 58.3 (Supported · #52/179)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 92.4% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 91.7% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 86.2% | Coming soon | Coming soon |

### Instruction Following

- Winner: Not comparable
- GPT-5.2 public-lane score: 92.4 (#24/157)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

### Mathematics

- Winner: Not comparable
- GPT-5.2 public-lane score: 65.3 (Unranked · 3 rankable rows)
- Qwen3-VL-235B-A22B Instruct public-lane score: Coming soon

| Benchmark | GPT-5.2 | Qwen3-VL-235B-A22B Instruct | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | 100% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 67.400% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 31.700% | Coming soon | Coming soon |

## FAQ

### Which is better overall, GPT-5.2 or Qwen3-VL-235B-A22B Instruct?

Qwen3-VL-235B-A22B Instruct is ahead overall on BenchLM right now.

### How many benchmarks does GPT-5.2 cover on BenchLM?

GPT-5.2 currently has 38 sourced benchmark scores on BenchLM.

### How many benchmarks does Qwen3-VL-235B-A22B Instruct cover on BenchLM?

Qwen3-VL-235B-A22B Instruct currently has 8 sourced benchmark scores on BenchLM.

## Related Comparisons

- [GPT-5.2 vs GPT-5.2 Pro](/compare/gpt-5-2-vs-gpt-5-2-pro)
- [Qwen3-VL-235B-A22B Instruct vs GPT-5.2 Pro](/compare/gpt-5-2-pro-vs-qwen3-vl-235b-a22b-instruct)
- [GPT-5.2 vs GPT-5.2 Instant](/compare/gpt-5-2-vs-gpt-5-2-instant)
- [Qwen3-VL-235B-A22B Instruct vs GPT-5.2 Instant](/compare/gpt-5-2-instant-vs-qwen3-vl-235b-a22b-instruct)

## Explore More

- [GPT-5.2 profile](/models/gpt-5-2)
- [Qwen3-VL-235B-A22B Instruct profile](/models/qwen3-vl-235b-a22b-instruct)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)
