Laguna S 2.1 vs MiniMax M3

Laguna S 2.1 and MiniMax M3 are closely matched at 41.4 and 41.9 on the LLM Stats Score. Laguna S 2.1 is 4.2x cheaper per token.

Poolside · MiniMax · Updated for 2026

Which is better?

Laguna S 2.1 and MiniMax M3 are closely matched on the overall LLM Stats Score at 41.4 and 41.9.

In the 3 individual benchmarks reported for both models, Laguna S 2.1 wins 3; this is a narrower head-to-head signal than the composite indexes.

On price, Laguna S 2.1 is roughly 4.2x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.

Laguna S 2.1 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 Laguna S 2.1

you value its reported benchmark strengths — it wins 3 of 3 exact shared results
cost matters — it's about 4.2x cheaper per token
you process long inputs — it offers a 1,048,576 token context window
you want the most recent training data — it shipped Jul 2026

Choose MiniMax M3

you want predictable pricing at $0.30/M input and $1.20/M output

At a glance

The differences that matter most.

Better token value: Laguna S 2.1

Core performance indexes

LLM Stats Score

41.4

#43 · 5 evals

41.9

#42 · 21 evals

Best

Reasoning

41.5

#40 · 4 evals

41.8

#39 · 17 evals

Best

Coding

33.1

#37 · 4 evals

34.4

#30 · 9 evals

Best

Agents

25.7

#43 · 4 evals

27.8

#38 · 14 evals

Best

Cost, coverage & limits

Benchmark wins

3 of 3

Best

0 of 3

Input price

$0.10 / M

Best

$0.30 / M

Output price

$0.20 / M

Best

$1.20 / M

Context window

1,048,576

Best

512,000

Capability indexes

Additional strengths measured across groups of related public benchmarks

1 shared

Index

Laguna S 2.1

MiniMax M3

Tool use

18.9#56 · 2 evals

22.8#42 · 3 evals

Conservative TrueSkill rating · higher is betterHow scores work

Individual benchmarks

6 reported for Laguna S 2.1 · 35 for MiniMax M3

3 shared

Laguna S 2.1 outperforms in 3 benchmarks (SWE Atlas - Codebase QnA, SWE-Bench Pro, Terminal-Bench 2.1), while MiniMax M3 is better at 0 benchmarks.

Laguna S 2.1 significantly outperforms across most benchmarks.

Fri Aug 28 2026 • llm-stats.com

Human preference

Blind head-to-head votes and playground preference scores

Pricing Analysis

Price comparison per million tokens

Laguna S 2.1 costs less

For input processing, Laguna S 2.1 ($0.10/1M tokens) is 3.0x cheaper than MiniMax M3 ($0.30/1M tokens).

For output processing, Laguna S 2.1 ($0.20/1M tokens) is 6.0x cheaper than MiniMax M3 ($1.20/1M tokens).

In conclusion, MiniMax M3 is more expensive than Laguna S 2.1.*

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

Lowest available price from all providers

Fri Aug 28 2026 • llm-stats.com

Laguna S 2.1

Input tokens$0.10

Output tokens$0.20

Best providerPoolside

MiniMax M3

Input tokens$0.30

Output tokens$1.20

Best providerFireworks

Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

310.0B diff

MiniMax M3 has 310.0B more parameters than Laguna S 2.1, making it 262.7% larger.

Laguna S 2.1

118.0Bparameters

MiniMax M3

428.0Bparameters

118.0B

Laguna S 2.1

428.0B

MiniMax M3

Context Window

Maximum input and output token capacity

Laguna S 2.1 accepts 1,048,576 input tokens compared to MiniMax M3's 512,000 tokens. Only MiniMax M3 specifies output context (131,072 tokens).

Laguna S 2.1

Input1,048,576 tokens

Output- tokens

MiniMax M3

Input512,000 tokens

Output131,072 tokens

Fri Aug 28 2026 • llm-stats.com

Input capabilities

Documented input modalities across available providers

MiniMax M3 supports multimodal inputs, whereas Laguna S 2.1 does not.

MiniMax M3 can handle both text and other forms of data like images, making it suitable for multimodal applications.

Laguna S 2.1

Text

Images

Audio

Video

MiniMax M3

Text

Images

Audio

Video

License

Usage and distribution terms

Laguna S 2.1 is licensed under OpenMDW License v1.1, while MiniMax M3 uses MIT.

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

Laguna S 2.1

OpenMDW License v1.1

Open weights

MiniMax M3

MIT

Open weights

Release Timeline

When each model was launched

Laguna S 2.1 was released on 2026-07-21, while MiniMax M3 was released on 2026-06-01.

Laguna S 2.1 is 2 months newer than MiniMax M3.

Laguna S 2.1

Jul 21, 2026

1 months ago

1mo newer

MiniMax M3

Jun 1, 2026

2 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

Provider Availability

Laguna S 2.1 is available from Poolside. MiniMax M3 is available from Fireworks, MiniMax, Novita, Together.

Laguna S 2.1

Poolside

Input Price:Input: $0.10/1MOutput Price:Output: $0.20/1M

MiniMax M3

Fireworks

Input Price:Input: $0.30/1MOutput Price:Output: $1.20/1M

MiniMax

Input Price:Input: $0.30/1MOutput Price:Output: $1.20/1M

Novita

Input Price:Input: $0.30/1MOutput Price:Output: $1.20/1M

Together

Input Price:Input: $0.30/1MOutput Price:Output: $1.20/1M

* 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 Laguna S 2.1 and MiniMax M3 side-by-side, then vote on the output you prefer.

Laguna S 2.1

✓ Preferred

MiniMax M3

Open in Playground

FAQ

Common questions about Laguna S 2.1 vs MiniMax M3.

Which is better, Laguna S 2.1 or MiniMax M3?

Laguna S 2.1 and MiniMax M3 are closely matched on the LLM Stats Score at 41.4 and 41.9. Laguna S 2.1 is made by Poolside and MiniMax M3 is made by MiniMax. The best choice depends on your use case — compare their capability indexes, individual benchmarks, pricing, and limits above.

How does Laguna S 2.1 compare to MiniMax M3 in benchmarks?

Laguna S 2.1 scores SWE-bench Multilingual: 78.5%, Terminal-Bench 2.1: 70.2%, SWE-Bench Pro: 59.4%, Toolathlon: 49.7%, SWE Atlas - Codebase QnA: 46.2%. MiniMax M3 scores OmniDocBench 1.5: 91.6%, SpreadSheetBench-v1: 89.3%, USAMO 2026: 85.7%, Video-MME: 85.4%, VideoMMMU: 84.6%.

Is Laguna S 2.1 cheaper than MiniMax M3?

Laguna S 2.1 is 3.0x cheaper for input tokens. Laguna S 2.1 costs $0.10/M input and $0.20/M output via poolside. MiniMax M3 costs $0.30/M input and $1.20/M output via fireworks.

What are the context window sizes for Laguna S 2.1 and MiniMax M3?

Laguna S 2.1 supports 1.0M tokens and MiniMax M3 supports 512K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.

What are the main differences between Laguna S 2.1 and MiniMax M3?

Key differences include LLM Stats Score (41.4 vs 41.9), context window (1.0M vs 512K), input pricing ($0.10 vs $0.30/M), multimodal support (no vs yes), licensing (OpenMDW License v1.1 vs MIT). See the full comparison above for benchmark-by-benchmark results.

Who makes Laguna S 2.1 and MiniMax M3?

Laguna S 2.1 is developed by Poolside and MiniMax M3 is developed by MiniMax.

Laguna S 2.1 vs MiniMax M3