The AI arena is free today

Open Superagent

Codestral-22B vs DeepSeek-V4-Flash-Vision-Exp

Comparing Codestral-22B and DeepSeek-V4-Flash-Vision-Exp across benchmarks, pricing, and capabilities.

Mistral AI · DeepSeek · Updated for 2026

Which is better?

Codestral-22B and DeepSeek-V4-Flash-Vision-Exp trade strengths across price, capabilities, and technical limits. The better choice depends on the workload.

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

Choose Codestral-22B

  • you need open weights you can self-host or fine-tune

Choose DeepSeek-V4-Flash-Vision-Exp

  • you want the most recent training data — it shipped Aug 2026

At a glance

The differences that matter most.

Benchmark wins
Input price
— / M
$0.22 / M
Output price
— / M
$0.66 / M
Context window
1,048,576
Released
May 2024
Aug 2026
License
MNPL-0.1
Unknown

Performance Benchmarks

Comparative analysis across standard metrics

No common benchmarks found

Codestral-22B and DeepSeek-V4-Flash-Vision-Expdon't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.

Arena Performance

Playground indexes and blind preference scores

Context Window

Maximum input and output token capacity

Only DeepSeek-V4-Flash-Vision-Exp specifies input context (1,048,576 tokens). Only DeepSeek-V4-Flash-Vision-Exp specifies output context (393,216 tokens).

Mistral AI
Codestral-22B
Input- tokens
Output- tokens
DeepSeek
DeepSeek-V4-Flash-Vision-Exp
Input1,048,576 tokens
Output393,216 tokens
Tue Aug 25 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

DeepSeek-V4-Flash-Vision-Exp supports multimodal inputs, whereas Codestral-22B does not.

DeepSeek-V4-Flash-Vision-Exp can handle both text and other forms of data like images, making it suitable for multimodal applications.

Codestral-22B

Text
Images
Audio
Video

DeepSeek-V4-Flash-Vision-Exp

Text
Images
Audio
Video

Release Timeline

When each model was launched

Codestral-22B was released on 2024-05-29, while DeepSeek-V4-Flash-Vision-Exp was released on 2026-08-21.

DeepSeek-V4-Flash-Vision-Exp is 27 months newer than Codestral-22B.

Codestral-22B

May 29, 2024

2.2 years ago

DeepSeek-V4-Flash-Vision-Exp

Aug 21, 2026

4 days ago

2.2yr 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

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion

Judge for yourself.

Run your own prompts against Codestral-22B and DeepSeek-V4-Flash-Vision-Exp side-by-side, then vote on the output you prefer.

Codestral-22B
✓ Preferred
DeepSeek-V4-Flash-Vision-Exp
Open in Playground

FAQ

Common questions about Codestral-22B vs DeepSeek-V4-Flash-Vision-Exp.

Which is better, Codestral-22B or DeepSeek-V4-Flash-Vision-Exp?

Codestral-22B (Mistral AI) and DeepSeek-V4-Flash-Vision-Exp (DeepSeek) each have strengths in different areas. Compare their benchmark scores, pricing, context windows, and capabilities above to determine which fits your needs.

How does Codestral-22B compare to DeepSeek-V4-Flash-Vision-Exp in benchmarks?

Codestral-22B scores HumanEvalFIM-Average: 91.6%, HumanEval: 81.1%, MBPP: 78.2%, Spider: 63.5%, HumanEval-Average: 61.5%. DeepSeek-V4-Flash-Vision-Exp scores Terminal-Bench 2.1: 83.9%, DSBench-Hard: 63.6%, DeepSWE: 59.3%, NL2Repo: 57.7%, ZEROBench: 35.0%.

What are the context window sizes for Codestral-22B and DeepSeek-V4-Flash-Vision-Exp?

Codestral-22B supports an unknown number of tokens and DeepSeek-V4-Flash-Vision-Exp supports 1.0M tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Codestral-22B and DeepSeek-V4-Flash-Vision-Exp?

Key differences include multimodal support (no vs yes), licensing (MNPL-0.1 vs Unknown). See the full comparison above for benchmark-by-benchmark results.

Who makes Codestral-22B and DeepSeek-V4-Flash-Vision-Exp?

Codestral-22B is developed by Mistral AI and DeepSeek-V4-Flash-Vision-Exp is developed by DeepSeek.