Mistral Large 2 vs Phi 4

Mistral Large 2 and Phi 4 are closely matched at 7.4 and 5.4 on the LLM Stats Score. Phi 4 is 34.3x cheaper per token.

Mistral AI · Microsoft · Updated for 2026

Which is better?

Mistral Large 2 and Phi 4 are closely matched on the overall LLM Stats Score at 7.4 and 5.4.

The models split the 2 individual benchmarks reported for both models evenly.

On price, Phi 4 is roughly 34.3x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Mistral Large 2 also accepts a larger context window (128,000 input tokens), making it the stronger choice for long documents and large codebases.

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

Choose Mistral Large 2

you process long inputs — it offers a 128,000 token context window

Choose Phi 4

cost matters — it's about 34.3x cheaper per token
you want the most recent training data — it shipped Dec 2024

At a glance

The differences that matter most.

Better token value: Phi 4

Core performance indexes

LLM Stats Score

7.4

#288 · 4 evals

Best

5.4

#296 · 13 evals

Reasoning

7.3

#281 · 4 evals

Best

6.5

#287 · 12 evals

Coding

13.0

#138 · 1 evals

Best

3.4

#217 · 1 evals

Cost, coverage & limits

Benchmark wins

1 of 2

Input price

$2.00 / M

$0.07 / M

Best

Output price

$6.00 / M

$0.14 / M

Best

Context window

128,000

Best

16,384

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared

Index

Mistral Large 2

Phi 4

Math

11.3#244 · 2 evals

12.8#230 · 7 evals

Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

5 reported for Mistral Large 2 · 13 for Phi 4

2 shared

Mistral Large 2 outperforms in 1 benchmarks (HumanEval), while Phi 4 is better at 1 benchmark (MMLU).

Both models are evenly matched across the benchmarks.

Fri Sep 18 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Phi 4 costs less

For input processing, Mistral Large 2 ($2.00/1M tokens) is 28.6x more expensive than Phi 4 ($0.07/1M tokens).

For output processing, Mistral Large 2 ($6.00/1M tokens) is 42.9x more expensive than Phi 4 ($0.14/1M tokens).

In conclusion, Mistral Large 2 is more expensive than Phi 4.*

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

Lowest available price from all providers

Fri Sep 18 2026 • llm-stats.com

Mistral Large 2

Input tokens$2.00

Output tokens$6.00

Best providerGoogle

Phi 4

Input tokens$0.07

Output tokens$0.14

Best providerDeepinfra

Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

108.3B diff

Mistral Large 2 has 108.3B more parameters than Phi 4, making it 736.7% larger.

Mistral Large 2

123.0Bparameters

Phi 4

14.7Bparameters

123.0B

Mistral Large 2

14.7B

Phi 4

Context Window

Maximum input and output token capacity

Mistral Large 2 accepts 128,000 input tokens compared to Phi 4's 16,384 tokens. Mistral Large 2 can generate longer responses up to 128,000 tokens, while Phi 4 is limited to 16,384 tokens.

Mistral Large 2

Input128,000 tokens

Output128,000 tokens

Phi 4

Input16,384 tokens

Output16,384 tokens

Fri Sep 18 2026 • llm-stats.com

License

Usage and distribution terms

Mistral Large 2 is licensed under Mistral Research License, while Phi 4 uses MIT.

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

Mistral Large 2

Mistral Research License

Open weights

Phi 4

MIT

Open weights

Release Timeline

When each model was launched

Mistral Large 2 was released on 2024-07-24, while Phi 4 was released on 2024-12-12.

Phi 4 is 5 months newer than Mistral Large 2.

Mistral Large 2

Jul 24, 2024

2.2 years ago

Phi 4

Dec 12, 2024

1.8 years ago

4mo newer

Knowledge Cutoff

When training data ends

Phi 4 has a documented knowledge cutoff of 2024-06-01, while Mistral Large 2's cutoff date is not specified.

We can confirm Phi 4's training data extends to 2024-06-01, but cannot make a direct comparison without Mistral Large 2's cutoff date.

Mistral Large 2

—

Phi 4

Jun 2024

Provider Availability

Mistral Large 2 is available from Google, Mistral AI. Phi 4 is available from DeepInfra.

Mistral Large 2

Google

Input Price:Input: $2.00/1MOutput Price:Output: $6.00/1M

Mistral

Input Price:Input: $2.00/1MOutput Price:Output: $6.00/1M

Phi 4

Deepinfra

Input Price:Input: $0.07/1MOutput Price:Output: $0.14/1M

* 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 Large 2 and Phi 4 side-by-side, then vote on the output you prefer.

Mistral Large 2

✓ Preferred

Phi 4

Open in Playground

FAQ

Common questions about Mistral Large 2 vs Phi 4.

Which is better, Mistral Large 2 or Phi 4?

Mistral Large 2 and Phi 4 are closely matched on the LLM Stats Score at 7.4 and 5.4. Mistral Large 2 is made by Mistral AI and Phi 4 is made by Microsoft. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Mistral Large 2 compare to Phi 4 in benchmarks?

Mistral Large 2 scores GSM8k: 93.0%, HumanEval: 92.0%, MT-Bench: 86.3%, MMLU: 84.0%, MMLU French: 82.8%. Phi 4 scores MMLU: 84.8%, HumanEval+: 82.8%, HumanEval: 82.6%, MGSM: 80.6%, MATH: 80.4%.

Is Mistral Large 2 cheaper than Phi 4?

Phi 4 is 28.6x cheaper for input tokens. Mistral Large 2 costs $2.00/M input and $6.00/M output via google. Phi 4 costs $0.07/M input and $0.14/M output via deepinfra.

What are the context window sizes for Mistral Large 2 and Phi 4?

Mistral Large 2 supports 128K tokens and Phi 4 supports 16K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Mistral Large 2 and Phi 4?

Key differences include LLM Stats Score (7.4 vs 5.4), context window (128K vs 16K), input pricing ($2.00 vs $0.07/M), licensing (Mistral Research License vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Mistral Large 2 and Phi 4?

Mistral Large 2 is developed by Mistral AI and Phi 4 is developed by Microsoft.

Mistral Large 2 vs Phi 4