The AI arena is free today

Open Superagent

Llama 3.2 90B Instruct vs LongCat-2.5

LongCat-2.5 leads the LLM Stats Score 50.0 to 4.7. Llama 3.2 90B Instruct is 1.4x cheaper per token.

Meta · Meituan · Updated for 2026

Which is better?

LongCat-2.5 leads the overall LLM Stats Score 50.0 to 4.7, ranking #20 overall.

On price, Llama 3.2 90B Instruct is roughly 1.4x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

LongCat-2.5 also accepts a larger context window (1,048,576 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 Llama 3.2 90B Instruct

  • cost matters — it's about 1.4x cheaper per token
  • you need open weights you can self-host or fine-tune

Choose LongCat-2.5

  • overall performance matters — it scores 50.0 and ranks #20 on LLM Stats
  • your work emphasizes reasoning — it leads those capability indexes
  • you process long inputs — it offers a 1,048,576 token context window
  • you want the most recent training data — it shipped Nov 2026

At a glance

The differences that matter most.

Llama 3.2 90B Instruct
Meta
LongCat-2.5
Meituan
Core performance indexes
LLM Stats Score
4.7
#338
50.0
#20
Reasoning
6.1
#325
48.3
#21
Cost, coverage & limits
Benchmark wins
—
—
Input price
$0.35 / M
$0.30 / M
Output price
$0.40 / M
$1.20 / M
Context window
128,000
1,048,576

Capability indexes

Additional strengths measured across groups of related public benchmarks

3 shared
Index
Llama 3.2 90B Instruct
LongCat-2.5
Math
11.4#259
37.2#37
Vision
2.6#195
32.5#29
Multimodal
5.2#160
34.8#15
Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

13 reported for Llama 3.2 90B Instruct · 23 for LongCat-2.5

No common benchmarks found

Llama 3.2 90B Instruct and LongCat-2.5don't have any common benchmark datasets to compare. They may have been evaluated on different testing suites.

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Llama 3.2 90B Instruct costs less

For input processing, Llama 3.2 90B Instruct ($0.35/1M tokens) is 1.2x more expensive than LongCat-2.5 ($0.30/1M tokens).

For output processing, Llama 3.2 90B Instruct ($0.40/1M tokens) is 3.0x cheaper than LongCat-2.5 ($1.20/1M tokens).

In conclusion, LongCat-2.5 is more expensive than Llama 3.2 90B Instruct.*

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

Lowest available price from all providers
Mon Nov 02 2026 • llm-stats.com
Meta
Llama 3.2 90B Instruct
Input tokens$0.35
Output tokens$0.40
Best providerDeepinfra
Meituan
LongCat-2.5
Input tokens$0.30
Output tokens$1.20
Best providerMeituan
Notice missing or incorrect data?

Model Size

Parameter count comparison

1510.0B diff

LongCat-2.5 has 1510.0B more parameters than Llama 3.2 90B Instruct, making it 1677.8% larger.

Meta
Llama 3.2 90B Instruct
90.0Bparameters
Meituan
LongCat-2.5
1.6Tparameters
90.0B
Llama 3.2 90B Instruct
1600.0B
LongCat-2.5

Context Window

Maximum input and output token capacity

LongCat-2.5 accepts 1,048,576 input tokens compared to Llama 3.2 90B Instruct's 128,000 tokens. LongCat-2.5 can generate longer responses up to 131,072 tokens, while Llama 3.2 90B Instruct is limited to 128,000 tokens.

Meta
Llama 3.2 90B Instruct
Input128,000 tokens
Output128,000 tokens
Meituan
LongCat-2.5
Input1,048,576 tokens
Output131,072 tokens
Mon Nov 02 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

Both Llama 3.2 90B Instruct and LongCat-2.5 support multimodal inputs.

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

Llama 3.2 90B Instruct

Text
Images
Audio
Video

LongCat-2.5

Text
Images
Audio
Video

Release Timeline

When each model was launched

Llama 3.2 90B Instruct was released on 2024-09-25, while LongCat-2.5 was released on 2026-11-01.

LongCat-2.5 is 26 months newer than Llama 3.2 90B Instruct.

Llama 3.2 90B Instruct

Sep 25, 2024

2.1 years ago

LongCat-2.5

Nov 1, 2026

-1 days ago

2.1yr 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

Provider Availability

Llama 3.2 90B Instruct is available from DeepInfra, Bedrock, Fireworks, Together, Hyperbolic. LongCat-2.5 is available from Meituan.

Llama 3.2 90B Instruct

deepinfra logo
Deepinfra
Input Price:Input: $0.35/1MOutput Price:Output: $0.40/1M
bedrock logo
AWS Bedrock
Input Price:Input: $0.72/1MOutput Price:Output: $0.72/1M
fireworks logo
Fireworks
Input Price:Input: $0.89/1MOutput Price:Output: $0.89/1M
together logo
Together
Input Price:Input: $1.20/1MOutput Price:Output: $1.20/1M
hyperbolic logo
Hyperbolic
Input Price:Input: $2.00/1MOutput Price:Output: $2.00/1M

LongCat-2.5

meituan logo
Meituan
Input Price:Input: $0.30/1MOutput Price:Output: $1.20/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?

Judge for yourself.

Run your own prompts against Llama 3.2 90B Instruct and LongCat-2.5 side-by-side, then vote on the output you prefer.

Llama 3.2 90B Instruct
✓ Preferred
LongCat-2.5
Open in Playground

FAQ

Common questions about Llama 3.2 90B Instruct vs LongCat-2.5.

Which is better, Llama 3.2 90B Instruct or LongCat-2.5?

LongCat-2.5 leads the LLM Stats Score 50.0 to 4.7. Llama 3.2 90B Instruct is made by Meta and LongCat-2.5 is made by Meituan. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Llama 3.2 90B Instruct compare to LongCat-2.5 in benchmarks?

Llama 3.2 90B Instruct scores AI2D: 92.3%, DocVQA: 90.1%, MGSM: 86.9%, MMLU: 86.0%, ChartQA: 85.5%. LongCat-2.5 scores MathVision: 96.5%, BrowseComp: 91.9%, CharXiv-R: 90.2%, Terminal-Bench 2.1: 88.4%, Video-MME: 85.2%.

Is Llama 3.2 90B Instruct cheaper than LongCat-2.5?

LongCat-2.5 is 1.2x cheaper for input tokens. Llama 3.2 90B Instruct costs $0.35/M input and $0.40/M output via deepinfra. LongCat-2.5 costs $0.30/M input and $1.20/M output via meituan.

What are the context window sizes for Llama 3.2 90B Instruct and LongCat-2.5?

Llama 3.2 90B Instruct supports 128K tokens and LongCat-2.5 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 Llama 3.2 90B Instruct and LongCat-2.5?

Key differences include LLM Stats Score (4.7 vs 50.0), context window (128K vs 1.0M), input pricing ($0.35 vs $0.30/M), licensing (Llama 3.2 vs Unknown). See the full comparison above for benchmark-by-benchmark results.

Who makes Llama 3.2 90B Instruct and LongCat-2.5?

Llama 3.2 90B Instruct is developed by Meta and LongCat-2.5 is developed by Meituan.

Related comparisons