Model Comparison

DeepSeek R1 Zero vs Llama 3.2 90B Instruct

DeepSeek R1 Zero significantly outperforms across most benchmarks.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

1 benchmarks

DeepSeek R1 Zero outperforms in 1 benchmarks (GPQA), while Llama 3.2 90B Instruct is better at 0 benchmarks.

DeepSeek R1 Zero significantly outperforms across most benchmarks.

Wed Apr 01 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Cost data unavailable.

Lowest available price from all providers
Wed Apr 01 2026 • llm-stats.com
DeepSeek
DeepSeek R1 Zero
Input tokens$0.00
Output tokens$0.00
Best providerUnknown Organization
Meta
Llama 3.2 90B Instruct
Input tokens$0.35
Output tokens$0.40
Best providerDeepinfra
Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

581.0B diff

DeepSeek R1 Zero has 581.0B more parameters than Llama 3.2 90B Instruct, making it 645.6% larger.

DeepSeek
DeepSeek R1 Zero
671.0Bparameters
Meta
Llama 3.2 90B Instruct
90.0Bparameters
671.0B
DeepSeek R1 Zero
90.0B
Llama 3.2 90B 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).

DeepSeek
DeepSeek R1 Zero
Input- tokens
Output- tokens
Meta
Llama 3.2 90B Instruct
Input128,000 tokens
Output128,000 tokens
Wed Apr 01 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Llama 3.2 90B Instruct supports multimodal inputs, whereas DeepSeek R1 Zero does not.

Llama 3.2 90B Instruct can handle both text and other forms of data like images, making it suitable for multimodal applications.

DeepSeek R1 Zero

Text
Images
Audio
Video

Llama 3.2 90B Instruct

Text
Images
Audio
Video

License

Usage and distribution terms

DeepSeek R1 Zero is licensed under MIT, while Llama 3.2 90B Instruct uses Llama 3.2.

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

DeepSeek R1 Zero

MIT

Open weights

Llama 3.2 90B Instruct

Llama 3.2

Open weights

Release Timeline

When each model was launched

DeepSeek R1 Zero was released on 2025-01-20, while Llama 3.2 90B Instruct was released on 2024-09-25.

DeepSeek R1 Zero is 4 months newer than Llama 3.2 90B Instruct.

DeepSeek R1 Zero

Jan 20, 2025

1.2 years ago

3mo newer
Llama 3.2 90B Instruct

Sep 25, 2024

1.5 years ago

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→

Key Takeaways

DeepSeek
Higher GPQA score (73.3% vs 46.7%)
Meta
Larger context window (128,000 tokens)
Supports multimodal inputs

Detailed Comparison

AI Model Comparison Table
Feature
DeepSeek
DeepSeek R1 Zero
Meta
Llama 3.2 90B Instruct

FAQ

Common questions about DeepSeek R1 Zero vs Llama 3.2 90B Instruct

DeepSeek R1 Zero significantly outperforms across most benchmarks. DeepSeek R1 Zero is made by DeepSeek and Llama 3.2 90B Instruct is made by Meta. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.
DeepSeek R1 Zero scores MATH-500: 95.9%, AIME 2024: 86.7%, GPQA: 73.3%, LiveCodeBench: 50.0%. Llama 3.2 90B Instruct scores AI2D: 92.3%, DocVQA: 90.1%, MGSM: 86.9%, MMLU: 86.0%, ChartQA: 85.5%.
DeepSeek R1 Zero supports an unknown number of tokens and Llama 3.2 90B Instruct supports 128K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.
Key differences include multimodal support (no vs yes), licensing (MIT vs Llama 3.2). See the full comparison above for benchmark-by-benchmark results.
DeepSeek R1 Zero is developed by DeepSeek and Llama 3.2 90B Instruct is developed by Meta.