ERNIE 4.5 300B-A47B
Baidu's first open release at frontier scale: a 300B MoE with 47B active parameters. Strongest open model for Chinese, with partial weight publication.
By Baidu · China
Updated 2026-09-15
When to pick this model
- Chinese-language production workloads at frontier quality
- Multilingual applications targeting East Asian markets
- Research benchmarking against Western open models
- Long-context analysis up to 128k tokens
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 180 GB |
| Q5_K_M | 215 GB |
| Q8_0 | 320 GB |
| FP16 (no quantization) | 600 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, ERNIE 4.5 300B-A47B is server-class even at Q4_K_M (180 GB). Stepping up to Q8_0 nearly doubles the footprint to 320 GB, and unquantized FP16 weights take 600 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, ERNIE 4.5 300B-A47B needs roughly 220 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 1 tokens/sec on entry-level GPUs, on the order of 5 tokens/sec on a mid-range card, and up to 15 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches ERNIE 4.5 300B-A47B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for ERNIE 4.5 300B-A47B |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 180 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 180 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Does not fit — needs 180 GB at Q4_K_M |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Does not fit — needs 180 GB at Q4_K_M |
| 32 GB | RTX 5090 | Does not fit — needs 180 GB at Q4_K_M |
Which hardware should you buy to run ERNIE 4.5 300B-A47B?
To run ERNIE 4.5 300B-A47B locally at Q4, you need ~180 GB for Q4 weights alone. Hardware option to compare: BOSGAME M5 128GB / 2TB (Ryzen AI Max+ 395). This model exceeds the practical GPU memory of this mini PC. Choose a smaller model or larger infrastructure.
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Strengths
- Best-in-class Chinese-language performance
- Efficient MoE inference with 47B active
- 300B total parameters at frontier scale
- 128k context window
Limitations
- Around 180 GB VRAM at Q4
- Baidu license has commercial restrictions
- Limited adoption and support outside China
- Only partial weights publicly released
Typical workloads
In our catalog grid, ERNIE 4.5 300B-A47B is filed under Open CN Frontier, Pro Chat — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multilingual workloads.
The 128k-token context window is large enough to hold entire codebases' worth of files or long reports in a single prompt, which is what makes local RAG and document analysis practical. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.
Architecture & training
Architecture: MoE · 300B · ERNIE 4.5 · Baidu PaddlePaddle
Training: Baidu — massive Chinese corpus + multilingual. Weights partially released.
The strongest open model for Chinese workloads, but licensing and limited ecosystem outside China constrain its reach.
Quick start
Install the runtime for your system: Windows, macOS or Linux. Check the exact model tag or GGUF quantization below; catalog IDs are not necessarily Ollama tags.
Start at 4096 tokens of context, then use ollama ps to check GPU/CPU placement. A default download may use a different quantization from the configurator’s memory estimate. Keep the free setup working before considering a kit.
# Déploiement complexe — vérifier la licence Baidu avant usageOr use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Your private ChatGPT, free, on your own machine in an hour — LM Studio, Ollama, Open WebUI, your documents, no cloud.
- Lifetime online access
- PDF + files
- 30-day refund
Similar models worth comparing
Frequently asked questions
How much VRAM does ERNIE 4.5 300B-A47B need?
At the recommended Q4_K_M quantization, ERNIE 4.5 300B-A47B needs about 180 GB of VRAM. Q8_0 takes 320 GB, and unquantized FP16 weights take 600 GB.
Can ERNIE 4.5 300B-A47B run without a GPU?
Yes — with roughly 220 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does ERNIE 4.5 300B-A47B support?
ERNIE 4.5 300B-A47B supports a 128k-token context window (131,072 tokens).
Can I use ERNIE 4.5 300B-A47B commercially?
Yes. ERNIE 4.5 300B-A47B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.
How fast is ERNIE 4.5 300B-A47B on consumer hardware?
Our compatibility engine estimates on the order of 5 tokens/sec on a mid-range GPU and up to 15 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of ERNIE 4.5 300B-A47B should I download first?
Start with Q4_K_M (180 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It does not fit a single 24 GB consumer card — plan for multi-GPU or server hardware.