Hy3 vs MiniMax M3
Hy3 and MiniMax M3 are closely matched at 42.9 and 41.2 on the LLM Stats Score. Hy3 is 1.9x cheaper per token.
Tencent · MiniMax · Updated for 2026
Which is better?
Hy3 and MiniMax M3 are closely matched on the overall LLM Stats Score at 42.9 and 41.2.
The models split the 10 individual benchmarks reported for both models evenly.
On price, Hy3 is roughly 1.9x cheaper per token on a blended 3:1 input/output basis, which adds up quickly at production volume.
MiniMax M3 also accepts a larger context window (524,288 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 Hy3
- cost matters — it's about 1.9x cheaper per token
- you want the most recent training data — it shipped Jul 2026
Choose MiniMax M3
- you process long inputs — it offers a 524,288 token context window
At a glance
The differences that matter most.
Capability indexes
Additional strengths measured across groups of related public benchmarks
Individual benchmarks
31 reported for Hy3 · 35 for MiniMax M3
Hy3 outperforms in 5 benchmarks (BrowseComp, CL-bench, MCP Atlas, NL2Repo, Terminal-Bench 2.1), while MiniMax M3 is better at 5 benchmarks (APEX-Agents, Claw-Eval, SWE-Bench Pro, SWE-Bench Verified, USAMO 2026).
Both models are evenly matched across the benchmarks.
Human preference
Blind head-to-head votes and playground preference scores
Pricing Analysis
Price comparison per million tokens
For input processing, Hy3 ($0.14/1M tokens) is 2.0x cheaper than MiniMax M3 ($0.28/1M tokens).
For output processing, Hy3 ($0.58/1M tokens) is 1.9x cheaper than MiniMax M3 ($1.10/1M tokens).
In conclusion, MiniMax M3 is more expensive than Hy3.*
* Using a 3:1 ratio of input to output tokens
Model Size
Parameter count comparison
MiniMax M3 has 133.0B more parameters than Hy3, making it 45.1% larger.
Context Window
Maximum input and output token capacity
MiniMax M3 accepts 524,288 input tokens compared to Hy3's 262,144 tokens. MiniMax M3 can generate longer responses up to 524,288 tokens, while Hy3 is limited to 262,144 tokens.
Input capabilities
Documented input modalities across available providers
MiniMax M3 supports multimodal inputs, whereas Hy3 does not.
MiniMax M3 can handle both text and other forms of data like images, making it suitable for multimodal applications.
Hy3
MiniMax M3
License
Usage and distribution terms
Hy3 is licensed under Apache 2.0, while MiniMax M3 uses MIT.
License differences may affect how you can use these models in commercial or open-source projects.
Apache 2.0
Open weights
MIT
Open weights
Release Timeline
When each model was launched
Hy3 was released on 2026-07-06, while MiniMax M3 was released on 2026-06-01.
Hy3 is 1 month newer than MiniMax M3.
Jul 6, 2026
2 months ago
1mo newerJun 1, 2026
3 months 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
Hy3 is available from DeepInfra. MiniMax M3 is available from DeepInfra, Fireworks, MiniMax, Novita, Together.
Hy3
MiniMax M3
Outputs Comparison
Judge for yourself.
Run your own prompts against Hy3 and MiniMax M3 side-by-side, then vote on the output you prefer.
FAQ
Common questions about Hy3 vs MiniMax M3.