The AI arena is free today

Open Superagent

Mistral Medium 3.5 vs o3

o3 shows notably better performance in the majority of benchmarks. Mistral Medium 3.5 is 1.2x cheaper per token.

Mistral AI · OpenAI · Updated for 2026

Which is better?

Mistral Medium 3.5 outperforms in 1 benchmarks (SWE-Bench Verified), while o3 is better at 3 benchmarks (AIME 2025, BrowseComp, COLLIE). o3 shows notably better performance in the majority of benchmarks.

On price, Mistral Medium 3.5 is roughly 1.2x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Mistral Medium 3.5 also accepts a larger context window (256,000 input tokens), making it the stronger choice for long documents and large codebases.

Based on current benchmark, pricing, and model metadata for 2026.

Choose Mistral Medium 3.5

  • cost matters — it's about 1.2x cheaper per token
  • you process long inputs — it offers a 256,000 token context window
  • you want the most recent training data — it shipped Apr 2026
  • you need open weights you can self-host or fine-tune

Choose o3

  • you want the strongest raw capability — it leads on 3 of 4 shared benchmarks

At a glance

The differences that matter most.

Benchmark wins
1 of 4
3 of 4
Input price
$1.50 / M
$2.00 / M
Output price
$7.50 / M
$8.00 / M
Context window
256,000
200,000
Released
Apr 2026
Apr 2025
License
Modified MIT License
Proprietary

Performance Benchmarks

Comparative analysis across standard metrics

4 benchmarks

Mistral Medium 3.5 outperforms in 1 benchmarks (SWE-Bench Verified), while o3 is better at 3 benchmarks (AIME 2025, BrowseComp, COLLIE).

o3 shows notably better performance in the majority of benchmarks.

Wed Aug 26 2026 • llm-stats.com

Arena Performance

Playground indexes and blind preference scores

Pricing Analysis

Price comparison per million tokens

Mistral Medium 3.5 costs less

For input processing, Mistral Medium 3.5 ($1.50/1M tokens) is 1.3x cheaper than o3 ($2.00/1M tokens).

For output processing, Mistral Medium 3.5 ($7.50/1M tokens) is 1.1x cheaper than o3 ($8.00/1M tokens).

In conclusion, o3 is more expensive than Mistral Medium 3.5.*

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

Lowest available price from all providers
Wed Aug 26 2026 • llm-stats.com
Mistral AI
Mistral Medium 3.5
Input tokens$1.50
Output tokens$7.50
Best providerMistral
OpenAI
o3
Input tokens$2.00
Output tokens$8.00
Best providerOpenAI
Notice missing or incorrect data?Start an Issue

Context Window

Maximum input and output token capacity

Mistral Medium 3.5 accepts 256,000 input tokens compared to o3's 200,000 tokens. Mistral Medium 3.5 can generate longer responses up to 256,000 tokens, while o3 is limited to 100,000 tokens.

Mistral AI
Mistral Medium 3.5
Input256,000 tokens
Output256,000 tokens
OpenAI
o3
Input200,000 tokens
Output100,000 tokens
Wed Aug 26 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Both Mistral Medium 3.5 and o3 support multimodal inputs.

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

Mistral Medium 3.5

Text
Images
Audio
Video

o3

Text
Images
Audio
Video

License

Usage and distribution terms

Mistral Medium 3.5 is licensed under Modified MIT License, while o3 uses a proprietary license.

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

Mistral Medium 3.5

Modified MIT License

Open weights

o3

Proprietary

Closed source

Release Timeline

When each model was launched

Mistral Medium 3.5 was released on 2026-04-29, while o3 was released on 2025-04-16.

Mistral Medium 3.5 is 13 months newer than o3.

Mistral Medium 3.5

Apr 29, 2026

3 months ago

1.0yr newer
o3

Apr 16, 2025

1.4 years ago

Knowledge Cutoff

When training data ends

o3 has a documented knowledge cutoff of 2024-05-31, while Mistral Medium 3.5's cutoff date is not specified.

We can confirm o3's training data extends to 2024-05-31, but cannot make a direct comparison without Mistral Medium 3.5's cutoff date.

Mistral Medium 3.5

o3

May 2024

Provider Availability

Mistral Medium 3.5 is available from Mistral AI. o3 is available from OpenAI.

Mistral Medium 3.5

mistral logo
Mistral
Input Price:Input: $1.50/1MOutput Price:Output: $7.50/1M

o3

openai logo
OpenAI
Input Price:Input: $2.00/1MOutput Price:Output: $8.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against Mistral Medium 3.5 and o3 side-by-side, then vote on the output you prefer.

Mistral Medium 3.5
✓ Preferred
o3
Open in Playground

FAQ

Common questions about Mistral Medium 3.5 vs o3.

Which is better, Mistral Medium 3.5 or o3?

o3 shows notably better performance in the majority of benchmarks. Mistral Medium 3.5 is made by Mistral AI and o3 is made by OpenAI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Mistral Medium 3.5 compare to o3 in benchmarks?

Mistral Medium 3.5 scores COLLIE: 95.8%, Tau3 Telecom: 91.4%, AIME 2025: 86.3%, SWE-Bench Verified: 77.6%, Tau3 Retail: 76.1%. o3 scores COLLIE: 98.4%, AIME 2024: 91.6%, ARC-AGI: 88.0%, MathVista: 86.8%, AIME 2025: 86.4%.

Is Mistral Medium 3.5 cheaper than o3?

Mistral Medium 3.5 is 1.3x cheaper for input tokens. Mistral Medium 3.5 costs $1.50/M input and $7.50/M output via mistral. o3 costs $2.00/M input and $8.00/M output via openai.

What are the context window sizes for Mistral Medium 3.5 and o3?

Mistral Medium 3.5 supports 256K tokens and o3 supports 200K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Mistral Medium 3.5 and o3?

Key differences include context window (256K vs 200K), input pricing ($1.50 vs $2.00/M), licensing (Modified MIT License vs Proprietary). See the full comparison above for benchmark-by-benchmark results.

Who makes Mistral Medium 3.5 and o3?

Mistral Medium 3.5 is developed by Mistral AI and o3 is developed by OpenAI.