The AI arena is free today

Open Superagent

Llama 3.2 90B Instruct vs Qwen2.5 VL 32B Instruct

Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct are closely matched at 5.2 and 9.7 on the LLM Stats Score.

Meta · Alibaba Cloud / Qwen Team · Updated for 2026

Which is better?

Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct are closely matched on the overall LLM Stats Score at 5.2 and 9.7.

In the 6 individual benchmarks reported for both models, Qwen2.5 VL 32B Instruct wins 4; this is a narrower head-to-head signal than the composite indexes.

Based on current LLM Stats indexes, shared benchmarks, pricing, and model metadata for 2026.

Choose Llama 3.2 90B Instruct

  • you want predictable pricing at $0.35/M input and $0.40/M output

Choose Qwen2.5 VL 32B Instruct

  • you value its reported benchmark strengths — it wins 4 of 6 exact shared results
  • you want the most recent training data — it shipped Feb 2025

At a glance

The differences that matter most.

Llama 3.2 90B Instruct
Meta
Qwen2.5 VL 32B Instruct
Alibaba Cloud / Qwen Team
Core performance indexes
LLM Stats Score
5.2
#296
9.7
#267
Reasoning
6.7
#284
8.4
#272
Cost, coverage & limits
Benchmark wins
2 of 6
4 of 6
Input price
$0.35 / M
— / M
Output price
$0.40 / M
— / M
Context window
128,000
—

Capability indexes

Additional strengths measured across groups of related public benchmarks

3 shared
Index
Llama 3.2 90B Instruct
Qwen2.5 VL 32B Instruct
Math
11.6#238
11.5#239
Vision
2.8#165
6.3#146
Multimodal
5.6#136
9.1#119
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

13 reported for Llama 3.2 90B Instruct · 28 for Qwen2.5 VL 32B Instruct

6 shared

Llama 3.2 90B Instruct outperforms in 2 benchmarks (GPQA, MMLU), while Qwen2.5 VL 32B Instruct is better at 4 benchmarks (DocVQA, MATH, MMMU, MMMU-Pro).

Qwen2.5 VL 32B Instruct shows notably better performance in the majority of benchmarks.

Mon Sep 14 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

56.5B diff

Llama 3.2 90B Instruct has 56.5B more parameters than Qwen2.5 VL 32B Instruct, making it 168.7% larger.

Meta
Llama 3.2 90B Instruct
90.0Bparameters
Alibaba Cloud / Qwen Team
Qwen2.5 VL 32B Instruct
33.5Bparameters
90.0B
Llama 3.2 90B Instruct
33.5B
Qwen2.5 VL 32B Instruct

Context Window

Maximum input and output token capacity

Only Llama 3.2 90B Instruct specifies input context (128,000 tokens). Only Llama 3.2 90B Instruct specifies output context (128,000 tokens).

Meta
Llama 3.2 90B Instruct
Input128,000 tokens
Output128,000 tokens
Alibaba Cloud / Qwen Team
Qwen2.5 VL 32B Instruct
Input- tokens
Output- tokens
Mon Sep 14 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Both Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct support multimodal inputs.

They are both capable of processing various types of data, offering versatility in application.

Llama 3.2 90B Instruct

Text
Images
Audio
Video

Qwen2.5 VL 32B Instruct

Text
Images
Audio
Video

License

Usage and distribution terms

Llama 3.2 90B Instruct is licensed under Llama 3.2, while Qwen2.5 VL 32B Instruct uses Apache 2.0.

License differences may affect how you can use these models in commercial or open-source projects.

Llama 3.2 90B Instruct

Llama 3.2

Open weights

Qwen2.5 VL 32B Instruct

Apache 2.0

Open weights

Release Timeline

When each model was launched

Llama 3.2 90B Instruct was released on 2024-09-25, while Qwen2.5 VL 32B Instruct was released on 2025-02-28.

Qwen2.5 VL 32B Instruct is 5 months newer than Llama 3.2 90B Instruct.

Llama 3.2 90B Instruct

Sep 25, 2024

2.0 years ago

Qwen2.5 VL 32B Instruct

Feb 28, 2025

1.5 years ago

5mo newer

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Judge for yourself.

Run your own prompts against Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct side-by-side, then vote on the output you prefer.

Llama 3.2 90B Instruct
✓ Preferred
Qwen2.5 VL 32B Instruct
Open in Playground

FAQ

Common questions about Llama 3.2 90B Instruct vs Qwen2.5 VL 32B Instruct.

Which is better, Llama 3.2 90B Instruct or Qwen2.5 VL 32B Instruct?

Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct are closely matched on the LLM Stats Score at 5.2 and 9.7. Llama 3.2 90B Instruct is made by Meta and Qwen2.5 VL 32B Instruct is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Llama 3.2 90B Instruct compare to Qwen2.5 VL 32B Instruct in benchmarks?

Llama 3.2 90B Instruct scores AI2D: 92.3%, DocVQA: 90.1%, MGSM: 86.9%, MMLU: 86.0%, ChartQA: 85.5%. Qwen2.5 VL 32B Instruct scores DocVQA: 94.8%, Android Control Low_EM: 93.3%, HumanEval: 91.5%, ScreenSpot: 88.5%, MBPP: 84.0%.

What are the context window sizes for Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct?

Llama 3.2 90B Instruct supports 128K tokens and Qwen2.5 VL 32B Instruct supports an unknown number of tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct?

Key differences include LLM Stats Score (5.2 vs 9.7), licensing (Llama 3.2 vs Apache 2.0). See the full comparison above for benchmark-by-benchmark results.

Who makes Llama 3.2 90B Instruct and Qwen2.5 VL 32B Instruct?

Llama 3.2 90B Instruct is developed by Meta and Qwen2.5 VL 32B Instruct is developed by Alibaba Cloud / Qwen Team.

Related comparisons