Model Comparison

Pixtral-12B vs Qwen3-235B-A22B-Instruct-2507

Qwen3-235B-A22B-Instruct-2507 significantly outperforms across most benchmarks. Pixtral-12B is 2.1x cheaper per token.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

1 benchmarks

Pixtral-12B outperforms in 0 benchmarks, while Qwen3-235B-A22B-Instruct-2507 is better at 1 benchmark (IFEval).

Qwen3-235B-A22B-Instruct-2507 significantly outperforms across most benchmarks.

Sat May 09 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Pixtral-12B costs less

For input processing, Pixtral-12B ($0.15/1M tokens) costs the same as Qwen3-235B-A22B-Instruct-2507 ($0.15/1M tokens).

For output processing, Pixtral-12B ($0.15/1M tokens) is 5.3x cheaper than Qwen3-235B-A22B-Instruct-2507 ($0.80/1M tokens).

In conclusion, Qwen3-235B-A22B-Instruct-2507 is more expensive than Pixtral-12B.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers

Sat May 09 2026 • llm-stats.com

Pixtral-12B

Input tokens$0.15

Output tokens$0.15

Best providerMistral

Qwen3-235B-A22B-Instruct-2507

Input tokens$0.15

Output tokens$0.80

Best providerFireworks

Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

222.6B diff

Qwen3-235B-A22B-Instruct-2507 has 222.6B more parameters than Pixtral-12B, making it 1795.2% larger.

Pixtral-12B

12.4Bparameters

Qwen3-235B-A22B-Instruct-2507

235.0Bparameters

12.4B

Pixtral-12B

235.0B

Qwen3-235B-A22B-Instruct-2507

Context Window

Maximum input and output token capacity

Qwen3-235B-A22B-Instruct-2507 accepts 262,144 input tokens compared to Pixtral-12B's 128,000 tokens. Qwen3-235B-A22B-Instruct-2507 can generate longer responses up to 131,072 tokens, while Pixtral-12B is limited to 8,192 tokens.

Pixtral-12B

Input128,000 tokens

Output8,192 tokens

Qwen3-235B-A22B-Instruct-2507

Input262,144 tokens

Output131,072 tokens

Sat May 09 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Pixtral-12B supports multimodal inputs, whereas Qwen3-235B-A22B-Instruct-2507 does not.

Pixtral-12B can handle both text and other forms of data like images, making it suitable for multimodal applications.

Pixtral-12B

Text

Images

Audio

Video

Qwen3-235B-A22B-Instruct-2507

Text

Images

Audio

Video

License

Usage and distribution terms

Both models are licensed under Apache 2.0.

Both models share the same licensing terms, providing consistent usage rights.

Pixtral-12B

Apache 2.0

Open weights

Qwen3-235B-A22B-Instruct-2507

Apache 2.0

Open weights

Release Timeline

When each model was launched

Pixtral-12B was released on 2024-09-17, while Qwen3-235B-A22B-Instruct-2507 was released on 2025-07-22.

Qwen3-235B-A22B-Instruct-2507 is 10 months newer than Pixtral-12B.

Pixtral-12B

Sep 17, 2024

1.6 years ago

Qwen3-235B-A22B-Instruct-2507

Jul 22, 2025

9 months ago

10mo 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

Provider Availability

Pixtral-12B is available from Mistral AI. Qwen3-235B-A22B-Instruct-2507 is available from Fireworks, Novita.

Pixtral-12B

Mistral

Input Price:Input: $0.15/1MOutput Price:Output: $0.15/1M

Qwen3-235B-A22B-Instruct-2507

Fireworks

Input Price:Input: $0.15/1MOutput Price:Output: $0.80/1M

Novita

Input Price:Input: $0.15/1MOutput Price:Output: $0.80/1M

* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

Pixtral-12B

View details

Mistral AI

Supports multimodal inputs

Less expensive output tokens

Qwen3-235B-A22B-Instruct-2507

View details

Alibaba Cloud / Qwen Team

Larger context window (262,144 tokens)

Higher IFEval score (88.7% vs 61.3%)

Detailed Comparison

AI Model Comparison Table
Feature	Pixtral-12B	Qwen3-235B-A22B-Instruct-2507

FAQ

Common questions about Pixtral-12B vs Qwen3-235B-A22B-Instruct-2507.

Which is better, Pixtral-12B or Qwen3-235B-A22B-Instruct-2507?

Qwen3-235B-A22B-Instruct-2507 significantly outperforms across most benchmarks. Pixtral-12B is made by Mistral AI and Qwen3-235B-A22B-Instruct-2507 is made by Alibaba Cloud / Qwen Team. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Pixtral-12B compare to Qwen3-235B-A22B-Instruct-2507 in benchmarks?

Pixtral-12B scores DocVQA: 90.7%, ChartQA: 81.8%, VQAv2: 78.6%, MT-Bench: 76.8%, HumanEval: 72.0%. Qwen3-235B-A22B-Instruct-2507 scores ZebraLogic: 95.0%, MMLU-Redux: 93.1%, IFEval: 88.7%, MultiPL-E: 87.9%, Creative Writing v3: 87.5%.

Is Pixtral-12B cheaper than Qwen3-235B-A22B-Instruct-2507?

Both models cost $0.15 per million input tokens.

What are the context window sizes for Pixtral-12B and Qwen3-235B-A22B-Instruct-2507?

Pixtral-12B supports 128K tokens and Qwen3-235B-A22B-Instruct-2507 supports 262K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Pixtral-12B and Qwen3-235B-A22B-Instruct-2507?

Key differences include context window (128K vs 262K), multimodal support (yes vs no). See the full comparison above for benchmark-by-benchmark results.

Who makes Pixtral-12B and Qwen3-235B-A22B-Instruct-2507?

Pixtral-12B is developed by Mistral AI and Qwen3-235B-A22B-Instruct-2507 is developed by Alibaba Cloud / Qwen Team.