Baidu · Released Jun 30, 2025ERNIE 4.5 21B A3B
Open ERNIE text MoE with 21B total and 3B active parameters.
- Capability
- —No ECI score yet
- Input
- $0.07per 1M tokens
- Output
- $0.28per 1M tokens
- Context
- 131K8K max output
What it costs
- Input
- $0.07per million tokens
- Output
- $0.28per million tokens
- Blended (3:1)
- $0.122cheaper than 87% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $0.28 |
| Team assistant25M input + 5M output tokens | $3.15 |
| Production app250M input + 50M output tokens | $31.50 |
Median of 1 paid API provider on models.dev.
Independent benchmarks
Epoch AI has not published a Capabilities Index score for ERNIE 4.5 21B A3B yet.
The details
- Lab
- Baidu
- Released
- Jun 30, 2025
- Context window
- 131,072 tokens
- Max output
- 8,000 tokens
- Inputs
- Text
- Output
- Text
- Reasoning
- No
- Tool calling
- Yes
- Structured output
- No
- Weights
- Open weights · Apache-2.0
- Availability
- 1 API providerlisted on models.dev
More from Baidu
About ERNIE 4.5 21B A3B
How much does ERNIE 4.5 21B A3B cost?
ERNIE 4.5 21B A3B costs $0.07 input / $0.28 output per million tokens (median across 1 API provider). At a 3:1 input-to-output mix that is $0.122 per million tokens, cheaper than 87% of the 445 priced models we track.
What is the context window of ERNIE 4.5 21B A3B?
ERNIE 4.5 21B A3B accepts up to 131,072 tokens per request and can write up to 8,000 tokens in one response.
How good is ERNIE 4.5 21B A3B?
Epoch AI has not published a Capabilities Index score for ERNIE 4.5 21B A3B yet.
Is ERNIE 4.5 21B A3B open source?
Yes. Baidu publishes the weights under the Apache-2.0 licence, so it can be downloaded and self-hosted.
What inputs does ERNIE 4.5 21B A3B support?
ERNIE 4.5 21B A3B accepts text and replies in text. It is not a dedicated reasoning model, supports tool calling.
When was ERNIE 4.5 21B A3B released?
Baidu released ERNIE 4.5 21B A3B on Jun 30, 2025.
What are the best alternatives to ERNIE 4.5 21B A3B?
The closest current models from other labs on capability, price and release date are Qwen Flash (Alibaba (Qwen), $0.05 / $0.40), Apertus 8B (Swiss AI, $0.10 / $0.20), Mistral Small 3.2 24B (Mistral AI, $0.10 / $0.305) and Nemotron Nano 9B v2 (NVIDIA, $0.06 / $0.23).