Decision support

ERNIE 4.5 VL 28B A3B Thinking vs MiniMax-M2

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

Benchmark results

  • Arena Text

    Arena rating · Higher is better

    • ERNIE 4.5 VL 28B A3B Thinking
      Not reported
    • MiniMax-M2
      1344
      Arena #232CI 1336 to 1351
    Runs and sources
    ERNIE 4.5 VL 28B A3B Thinking

    Not reported

    MiniMax-M2
    • default134495% CI 1336 to 1351 · 6,832 votes
  • Arena WebDev

    Arena rating · Higher is better

    • ERNIE 4.5 VL 28B A3B Thinking
      Not reported
    • MiniMax-M2
      1299
      Arena #136CI 1288 to 1310
    Runs and sources
    ERNIE 4.5 VL 28B A3B Thinking

    Not reported

    MiniMax-M2
    • default129995% CI 1288 to 1310 · 6,599 votes
  • Terminal Bench

    Best published score · Higher is better

    • ERNIE 4.5 VL 28B A3B Thinking
      Not reported
    • MiniMax-M2
      30.0%
      #38 of 49
    Runs and sources
    ERNIE 4.5 VL 28B A3B Thinking

    Not reported

    MiniMax-M2
    • default30.0%Terminus 2

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

Original-source terms apply to evaluations credited to their original publishers.

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

Model specifications

Release date
November 11, 2025
October 27, 2025
Context window
—
205K
Native context
—
—
Max output tokens
—
131,072
Knowledge cutoff
—
—
Open weights
Open weights
Open weights
Parameter count
29.7B
229B

API pricing · USD per 1M tokens

Input
—
—
Output
—
—
Cached input
—
—
Cache write
—
—

Capabilities and modalities

Input modalities
text, image
text
Output modalities
text
text
Tool calling
—
Supported
Reasoning option
—
Supported
Supported options
—
tools, reasoning, temperature
Capabilities
—
—

Architecture and availability

Architecture
—
—
Model families
—
—
Tokenizer
—
—
Instruction format
—
—
License
Apache 2.0
—
Providers
0
0
Provider endpoints
0
0
Ollama variants
0
0
Metadata source
curated
models.dev

A dash means the field is not reported. Provider-specific performance and pricing may differ.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.