Decision support

ERNIE 4.5 VL 424B A47B vs Qwen3 VL 235B A22B Instruct

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

Models Hardware

ERNIE 4.5 VL 424B A47B

Baidu

Sources:models.devOpenRouter

Qwen3 VL 235B A22B Instruct

Qwen

Sources:models.devOpenRouter

Benchmark results

Arena Text
Arena rating · Higher is better
- ERNIE 4.5 VL 424B A47B
  Not reported
- Qwen3 VL 235B A22B Instruct
  1413
  Arena #150CI 1407 to 1420
Runs and sources
ERNIE 4.5 VL 424B A47B
Not reported
Qwen3 VL 235B A22B Instruct
default141395% CI 1407 to 1420 · 11,288 votes
Source:Arena Text leaderboard, 2026-11-09
Arena Vision
Arena rating · Higher is better
- ERNIE 4.5 VL 424B A47B
  Not reported
- Qwen3 VL 235B A22B Instruct
  1214
  Arena #84CI 1208 to 1221
Runs and sources
ERNIE 4.5 VL 424B A47B
Not reported
Qwen3 VL 235B A22B Instruct
default121495% CI 1208 to 1221 · 12,672 votes
Source:Arena Vision leaderboard, 2026-11-09

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.

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

June 30, 2025

September 23, 2025

Context window

123K

262K

Native context

—

256K

Max output tokens

16,000

32,768

Knowledge cutoff

2025-03-31

Open weights

Parameter count

424B

235B

API pricing · USD per 1M tokens

Input

$0.42

$0.21

Output

$1.25

$1.90

Cached input

—

$0.10

Cache write

—

Capabilities and modalities

Input modalities

image, text

text, image

Output modalities

text

Tool calling

—

Supported

Reasoning option

Supported

—

Supported options

frequency_penalty, include_reasoning, max_tokens, presence_penalty, reasoning, repetition_penalty, seed, stop, temperature, top_k, top_p

frequency_penalty, logit_bias, logprobs, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

Capabilities

—

Architecture and availability

Architecture

—

Model families

—

Tokenizer

Other

Qwen3

Instruction format

—

License

Apache 2.0

—

Providers

Provider endpoints

Ollama variants

Metadata source

ai-research

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

Benchmark results

Arena Text

Arena Vision

Model specifications

API pricing · USD per 1M tokens

Capabilities and modalities

Architecture and availability