The AI arena is free today

Open Superagent

Command A+ vs LongCat-Flash-Thinking

Command A+ and LongCat-Flash-Thinking are closely matched at 24.0 and 28.6 on the LLM Stats Score.

Cohere · Meituan · Updated for 2026

Which is better?

Command A+ and LongCat-Flash-Thinking are closely matched on the overall LLM Stats Score at 24.0 and 28.6.

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

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

Choose Command A+

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

Choose LongCat-Flash-Thinking

  • your work emphasizes reasoning — it leads those capability indexes

At a glance

The differences that matter most.

Command A+
Cohere
LongCat-Flash-Thinking
Meituan
Core performance indexes
LLM Stats Score
24.0
#171
28.6
#136
Reasoning
21.5
#182
28.8
#127
Coding
7.3
#188
14.0
#138
Cost, coverage & limits
Benchmark wins
1 of 2
1 of 2
Input price
— / M
$0.30 / M
Output price
— / M
$1.20 / M
Context window
—
128,000

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared
Index
Command A+
LongCat-Flash-Thinking
Math
24.1#119
28.1#95
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

15 reported for Command A+ · 14 for LongCat-Flash-Thinking

2 shared

Command A+ outperforms in 1 benchmarks (Tau2 Telecom), while LongCat-Flash-Thinking is better at 1 benchmark (AIME 2025).

Both models are evenly matched across the benchmarks.

Tue Sep 22 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Model Size

Parameter count comparison

342.0B diff

LongCat-Flash-Thinking has 342.0B more parameters than Command A+, making it 156.9% larger.

Cohere
Command A+
218.0Bparameters
Meituan
LongCat-Flash-Thinking
560.0Bparameters
218.0B
Command A+
560.0B
LongCat-Flash-Thinking

Context Window

Maximum input and output token capacity

Only LongCat-Flash-Thinking specifies input context (128,000 tokens). Only LongCat-Flash-Thinking specifies output context (128,000 tokens).

Cohere
Command A+
Input- tokens
Output- tokens
Meituan
LongCat-Flash-Thinking
Input128,000 tokens
Output128,000 tokens
Tue Sep 22 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Command A+ supports multimodal inputs, whereas LongCat-Flash-Thinking does not.

Command A+ can handle both text and other forms of data like images, making it suitable for multimodal applications.

Command A+

Text
Images
Audio
Video

LongCat-Flash-Thinking

Text
Images
Audio
Video

License

Usage and distribution terms

Command A+ is licensed under Apache 2.0, while LongCat-Flash-Thinking uses MIT.

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

Command A+

Apache 2.0

Open weights

LongCat-Flash-Thinking

MIT

Open weights

Release Timeline

When each model was launched

Command A+ was released on 2026-05-20, while LongCat-Flash-Thinking was released on 2025-09-22.

Command A+ is 8 months newer than LongCat-Flash-Thinking.

Command A+

May 20, 2026

4 months ago

8mo newer
LongCat-Flash-Thinking

Sep 22, 2025

1.0 years 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 Command A+ and LongCat-Flash-Thinking side-by-side, then vote on the output you prefer.

Command A+
✓ Preferred
LongCat-Flash-Thinking
Open in Playground

FAQ

Common questions about Command A+ vs LongCat-Flash-Thinking.

Which is better, Command A+ or LongCat-Flash-Thinking?

Command A+ and LongCat-Flash-Thinking are closely matched on the LLM Stats Score at 24.0 and 28.6. Command A+ is made by Cohere and LongCat-Flash-Thinking is made by Meituan. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Command A+ compare to LongCat-Flash-Thinking in benchmarks?

Command A+ scores AIME 2025: 90.0%, CharXiv-D: 88.0%, MT-AIME 2025: 86.0%, Tau2 Telecom: 85.0%, WMT24++: 81.0%. LongCat-Flash-Thinking scores MATH-500: 99.2%, ZebraLogic: 95.5%, AIME 2024: 93.3%, AIME 2025: 90.6%, MMLU-Redux: 89.3%.

What are the context window sizes for Command A+ and LongCat-Flash-Thinking?

Command A+ supports an unknown number of tokens and LongCat-Flash-Thinking supports 128K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Command A+ and LongCat-Flash-Thinking?

Key differences include LLM Stats Score (24.0 vs 28.6), multimodal support (yes vs no), licensing (Apache 2.0 vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Command A+ and LongCat-Flash-Thinking?

Command A+ is developed by Cohere and LongCat-Flash-Thinking is developed by Meituan.

Related comparisons