MiniMax M2 vs Qwen3-235B-A22B-Instruct-2507
MiniMax M2 and Qwen3-235B-A22B-Instruct-2507 are closely matched at 26.8 and 24.2 on the LLM Stats Score. Qwen3-235B-A22B-Instruct-2507 is 2.6x cheaper per token.
MiniMax · Alibaba Cloud / Qwen Team · Updated for 2026
Which is better?
MiniMax M2 and Qwen3-235B-A22B-Instruct-2507 are closely matched on the overall LLM Stats Score at 26.8 and 24.2.
In the 3 individual benchmarks reported for both models, MiniMax M2 wins 2; this is a narrower head-to-head signal than the composite indexes.
On price, Qwen3-235B-A22B-Instruct-2507 is roughly 2.6x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
MiniMax M2 also accepts a larger context window (1,000,000 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 MiniMax M2
- you value its reported benchmark strengths — it wins 2 of 3 exact shared results
- you process long inputs — it offers a 1,000,000 token context window
- you want the most recent training data — it shipped Oct 2025
Choose Qwen3-235B-A22B-Instruct-2507
- cost matters — it's about 2.6x cheaper per token
At a glance
The differences that matter most.
Capability indexes
Additional strengths measured across groups of related public benchmarks
Individual benchmarks
16 reported for MiniMax M2 · 25 for Qwen3-235B-A22B-Instruct-2507
MiniMax M2 outperforms in 2 benchmarks (AIME 2025, GPQA), while Qwen3-235B-A22B-Instruct-2507 is better at 1 benchmark (MMLU-Pro).
MiniMax M2 shows notably better performance in the majority of benchmarks.
Human preference
Blind head-to-head votes and playground preference scores
Pricing Analysis
Price comparison per million tokens
For input processing, MiniMax M2 ($0.30/1M tokens) is 3.3x more expensive than Qwen3-235B-A22B-Instruct-2507 ($0.09/1M tokens).
For output processing, MiniMax M2 ($1.20/1M tokens) is 2.2x more expensive than Qwen3-235B-A22B-Instruct-2507 ($0.55/1M tokens).
In conclusion, MiniMax M2 is more expensive than Qwen3-235B-A22B-Instruct-2507.*
* Using a 3:1 ratio of input to output tokens
Model Size
Parameter count comparison
Qwen3-235B-A22B-Instruct-2507 has 5.0B more parameters than MiniMax M2, making it 2.2% larger.
Context Window
Maximum input and output token capacity
MiniMax M2 accepts 1,000,000 input tokens compared to Qwen3-235B-A22B-Instruct-2507's 262,144 tokens. MiniMax M2 can generate longer responses up to 1,000,000 tokens, while Qwen3-235B-A22B-Instruct-2507 is limited to 262,144 tokens.
License
Usage and distribution terms
MiniMax M2 is licensed under MIT, while Qwen3-235B-A22B-Instruct-2507 uses Apache 2.0.
License differences may affect how you can use these models in commercial or open-source projects.
MIT
Open weights
Apache 2.0
Open weights
Release Timeline
When each model was launched
MiniMax M2 was released on 2025-10-27, while Qwen3-235B-A22B-Instruct-2507 was released on 2025-07-22.
MiniMax M2 is 3 months newer than Qwen3-235B-A22B-Instruct-2507.
Oct 27, 2025
11 months ago
3mo newerJul 22, 2025
1.2 years ago
Knowledge Cutoff
When training data ends
Neither model specifies a knowledge cutoff date.
Unable to compare the recency of their training data.
Provider Availability
MiniMax M2 is available from MiniMax, Novita. Qwen3-235B-A22B-Instruct-2507 is available from DeepInfra, Fireworks, Novita.
MiniMax M2
Qwen3-235B-A22B-Instruct-2507
Outputs Comparison
Judge for yourself.
Run your own prompts against MiniMax M2 and Qwen3-235B-A22B-Instruct-2507 side-by-side, then vote on the output you prefer.
FAQ
Common questions about MiniMax M2 vs Qwen3-235B-A22B-Instruct-2507.