ERNIE 4.5 21B A3B Thinking
ERNIE 4.5 21B A3B Thinking is Baidu's language model with a 131K context window and up to 66K output tokens, starting at $0.07 / 1M input and $0.28 / 1M output. A reasoning-focused MoE LLM from Baidu with 21B total and 3B active parameters, enhanced for complex multi-step reasoning tasks.
Capabilities
Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities1/13
Reasoning✓
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | |
|---|---|---|
| Input $ / 1M | Output $ / 1M | |
| $0.07 | $0.28 | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host ERNIE 4.5 21B A3B Thinking, ranked by cheapest on-demand price. The model needs about 50 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
Instance | Cloud | GPU | VRAM | Price | Cheapest region | |
|---|---|---|---|---|---|---|
| Standard_NP10s |  Azure | AMD Alveo U250 FPGA (64GB) | 64 GB | $1.65/hr | westus2 | |
| Standard_NV32as_v4 | Azure | AMD Radeon Instinct MI25 | 64 GB | $1.86/hr | westus2 | |
| Standard_NP20s | Azure | 2× AMD Alveo U250 FPGA (64GB) | 128 GB | $3.30/hr | westus2 | |
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| ERNIE 5 Thinking Preview | — | — | — | — | Available |
| ERNIE 4.5 21B A3B Thinking | 131K | $0.070 | $0.280 | Current | |
| ERNIE 4.5 300B A47B | 131K | — | — | Available | |
| ERNIE 4.5 VL 424B A47B | 131K | $0.420 | $1.25 | Available | |
| ERNIE 4.5 300B A47B Paddle | — | 123K | $0.280 | $1.10 | Available |
| ERNIE 4.5 VL 28B A3B Thinking | — | 131K | $0.390 | $0.390 | Available |
| ERNIE Image | — | — | — | — | Available |
| ERNIE Image Turbo | — | — | — | — | Available |