The AI arena is free today

Open Superagent

Mistral Large 3 (675B Base) vs Mistral Large 3 (675B Instruct 2512 Eagle)

Both models are evenly matched across the benchmarks.

Mistral AI · Mistral AI · Updated for 2026

Which is better?

Mistral Large 3 (675B Base) outperforms in 0 benchmarks, while Mistral Large 3 (675B Instruct 2512 Eagle) is better at 0 benchmarks. Both models are evenly matched across the benchmarks.

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

Choose Mistral Large 3 (675B Base)

  • you are already invested in the Mistral AI ecosystem

Choose Mistral Large 3 (675B Instruct 2512 Eagle)

  • you want the strongest raw capability — it leads on 5 of 5 shared benchmarks

At a glance

The differences that matter most.

Mistral Large 3 (675B Base)
Mistral AI
Mistral Large 3 (675B Instruct 2512 Eagle)
Mistral AI
Benchmark wins
0 of 5
5 of 5
Input price
— / M
— / M
Output price
— / M
— / M
Context window
—
—
Released
Dec 2025
Dec 2025
License
Apache 2.0
Apache 2.0

Performance Benchmarks

Comparative analysis across standard metrics

5 benchmarks

Mistral Large 3 (675B Base) outperforms in 0 benchmarks, while Mistral Large 3 (675B Instruct 2512 Eagle) is better at 0 benchmarks.

Both models are evenly matched across the benchmarks.

Mon Aug 24 2026 • llm-stats.com

Arena Performance

Playground indexes and blind preference scores

Model Size

Parameter count comparison

0.0M diff

Mistral Large 3 (675B Instruct 2512 Eagle) has 0.0B more parameters than Mistral Large 3 (675B Base), making it 0.0% larger.

Mistral AI
Mistral Large 3 (675B Base)
675.0Bparameters
Mistral AI
Mistral Large 3 (675B Instruct 2512 Eagle)
675.0Bparameters
675.0B
Mistral Large 3 (675B Base)
675.0B
Mistral Large 3 (675B Instruct 2512 Eagle)

Input Capabilities

Supported data types and modalities

Both Mistral Large 3 (675B Base) and Mistral Large 3 (675B Instruct 2512 Eagle) support multimodal inputs.

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

Mistral Large 3 (675B Base)

Text
Images
Audio
Video

Mistral Large 3 (675B Instruct 2512 Eagle)

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.

Mistral Large 3 (675B Base)

Apache 2.0

Open weights

Mistral Large 3 (675B Instruct 2512 Eagle)

Apache 2.0

Open weights

Release Timeline

When each model was launched

Both models were released on 2025-12-04.

They likely represent similar generations of model development.

Mistral Large 3 (675B Base)

Dec 4, 2025

8 months ago

Mistral Large 3 (675B Instruct 2512 Eagle)

Dec 4, 2025

8 months 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→

Judge for yourself.

Run your own prompts against Mistral Large 3 (675B Base) and Mistral Large 3 (675B Instruct 2512 Eagle) side-by-side, then vote on the output you prefer.

Mistral Large 3 (675B Base)
✓ Preferred
Mistral Large 3 (675B Instruct 2512 Eagle)
Open in Playground

FAQ

Common questions about Mistral Large 3 (675B Base) vs Mistral Large 3 (675B Instruct 2512 Eagle).

Which is better, Mistral Large 3 (675B Base) or Mistral Large 3 (675B Instruct 2512 Eagle)?

Both models are evenly matched across the benchmarks. Mistral Large 3 (675B Base) is made by Mistral AI and Mistral Large 3 (675B Instruct 2512 Eagle) is made by Mistral AI. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.

How does Mistral Large 3 (675B Base) compare to Mistral Large 3 (675B Instruct 2512 Eagle) in benchmarks?

Mistral Large 3 (675B Base) scores MMMLU: 85.5%, AMC_2022_23: 52.0%, GPQA: 43.9%, LiveCodeBench: 34.4%, SimpleQA: 23.8%. Mistral Large 3 (675B Instruct 2512 Eagle) scores MMMLU: 85.5%, AMC_2022_23: 52.0%, GPQA: 43.9%, LiveCodeBench: 34.4%, SimpleQA: 23.8%.

Related comparisons