Can NVIDIA GeForce RTX 4090 run ERNIE 4.5 21B A3B PT?

Yes, at Q8_0 (21.4 GB) or lower. Higher quantizations like BF16 (42.4 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for ERNIE 4.5 21B A3B PT?

For ERNIE 4.5 21B A3B PT, Q4_K_M (13.0 GB) offers the best balance of quality and VRAM usage. Q4_K_L (13.3 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 6.2 GB.

Can I run ERNIE 4.5 21B A3B PT on a Mac?

ERNIE 4.5 21B A3B PT requires at least 6.2 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.

Can I run ERNIE 4.5 21B A3B PT locally?

Yes — ERNIE 4.5 21B A3B PT can run locally on consumer hardware. At Q4_K_M quantization it needs 13.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is ERNIE 4.5 21B A3B PT?

At Q4_K_M, ERNIE 4.5 21B A3B PT can reach ~338 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~50 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

What's the download size of ERNIE 4.5 21B A3B PT?

At Q4_K_M, the download is about 12.60 GB. The full-precision BF16 version is 42.00 GB. The smallest option (IQ2_XXS) is 5.78 GB.

Baidu·ERNIE·Ernie4_5_MoeForCausalLM

ERNIE 4.5 21B A3B PT — Hardware Requirements & GPU Compatibility

Chat

ERNIE 4.5 21B A3B PT is a 21B-parameter open language model from Baidu in the ERNIE family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 13.02 GB of VRAM — see which GPUs and Macs can run it below.

25.6K downloads 165 likes 2.5K quant downloadsJun 2025131K context

Specifications

Publisher: Baidu
Family: ERNIE
Parameters: 21B
Architecture: Ernie4_5_MoeForCausalLM
Context Length: 131,072 tokens
Vocabulary Size: 103,424
Release Date: 2025-06-28
License: Apache 2.0

Get Started

HuggingFace

baidu/ERNIE-4.5-21B-A3B-PT

How Much VRAM Does ERNIE 4.5 21B A3B PT Need?

Select a quantization to see compatible GPUs below.

Quantization	Bits	VRAM	+ Context	File Size	Quality
Q2_K	3.40	9.3 GB	16.7 GB	8.93 GB	2-bit quantization with K-quant improvements
Q3_K_S	3.50	9.6 GB	17 GB	9.19 GB	3-bit small quantization
Q3_K_M	3.90	10.7 GB	18.1 GB	10.24 GB	3-bit medium quantization
Q4_0	4.00	10.9 GB	18.3 GB	10.50 GB	4-bit legacy quantization
Q4_K_M	4.80	13.0 GB	20.4 GB	12.60 GB	4-bit medium quantization — most popular sweet spot
Q5_K_M	5.70	15.4 GB	22.8 GB	14.96 GB	5-bit medium quantization — good quality/size tradeoff
Q6_K	6.60	17.7 GB	25.1 GB	17.32 GB	6-bit quantization, very good quality
Q8_0	8.00	21.4 GB	28.8 GB	21.00 GB	8-bit quantization, near-lossless

Which GPUs Can Run ERNIE 4.5 21B A3B PT?

Q4_K_M · 13.0 GB

Show professional

ERNIE 4.5 21B A3B PT (Q4_K_M) requires 13.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 17+ GB is recommended. Using the full 131K context window can add up to 7.4 GB, bringing total usage to 20.4 GB. 26 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 5080.

Runs great

— Plenty of headroom

NVIDIA GeForce RTX 5090~90 tok/s NVIDIA GeForce RTX 3090 Ti~50 tok/s NVIDIA GeForce RTX 4090~50 tok/s NVIDIA GeForce RTX 3090~47 tok/s NVIDIA GeForce RTX 5090 Laptop GPU~45 tok/s AMD Radeon RX 7900 XTX~41 tok/s AMD Radeon RX 7900 XT~34 tok/s NVIDIA TITAN RTX~34 tok/s

Decent

— Enough VRAM, may be tight

NVIDIA GeForce RTX 5080~48 tok/s NVIDIA GeForce RTX 5070 Ti~45 tok/s NVIDIA GeForce RTX 4080 SUPER~37 tok/s NVIDIA GeForce RTX 4080~36 tok/s NVIDIA GeForce RTX 4070 Ti SUPER~34 tok/s NVIDIA GeForce RTX 4090 Laptop GPU~29 tok/s AMD Radeon RX 9070~27 tok/s AMD Radeon RX 9070 XT~27 tok/s AMD Radeon RX 7800 XT~26 tok/s AMD Radeon RX 7900 GRE~24 tok/s NVIDIA GeForce RTX 5060 Ti 16GB~22 tok/s AMD Radeon RX 6800~22 tok/s AMD Radeon RX 6800 XT~22 tok/s AMD Radeon RX 6900 XT~22 tok/s Intel Arc A770 16GB~22 tok/s NVIDIA GeForce RTX 4060 Ti 16GB~14 tok/s AMD Radeon RX 9060 XT 16GB~14 tok/s AMD Radeon RX 7600 XT~12 tok/s

Which Devices Can Run ERNIE 4.5 21B A3B PT?

Q4_K_M · 13.0 GB

47 devices with unified memory can run ERNIE 4.5 21B A3B PT, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 (16 GB).

Runs great

— Plenty of headroom

Decent

— Enough memory, may be tight

Mac Mini M4 (16 GB)~7 tok/s MacBook Air 13" M4 (16 GB)~7 tok/s MacBook Air 15" M4 (16 GB)~7 tok/s MacBook Pro 14" M4 (16 GB)~7 tok/s iPad Pro M4 13" (16 GB)~7 tok/s MacBook Air 13" M3 (16 GB)~6 tok/s

Where to Download ERNIE 4.5 21B A3B PT

Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.

unsloth/ERNIE-4.5-21B-A3B-PT-GGUF · 17 GGUF quants · 2.0K downloads bartowski/baidu_ERNIE-4.5-21B-A3B-PT-GGUF · 23 GGUF quants · 465 downloads

Related Models

ERNIE family · all 4 models by size →

Frequently Asked Questions

How much VRAM does ERNIE 4.5 21B A3B PT need?: ERNIE 4.5 21B A3B PT requires 13.0 GB of VRAM at Q4_K_M, or 42.4 GB at BF16. Full 131K context adds up to 7.4 GB (20.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 21B × 4.8 bits ÷ 8 = 12.6 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 7.8 GB (at full 131K context)
VRAM usage by quantization
Q4_K_M
13.0 GB
Q4_K_M + full context
20.4 GB
Learn more about VRAM estimation →
Can NVIDIA GeForce RTX 4090 run ERNIE 4.5 21B A3B PT?: Yes, at Q8_0 (21.4 GB) or lower. Higher quantizations like BF16 (42.4 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for ERNIE 4.5 21B A3B PT?: For ERNIE 4.5 21B A3B PT, Q4_K_M (13.0 GB) offers the best balance of quality and VRAM usage. Q4_K_L (13.3 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 6.2 GB.
VRAM requirement by quantization
IQ2_XXS
6.2 GB~53%
Q2_K
9.3 GB~75%
IQ4_XS
11.7 GB~87%
Q4_K_M ★
13.0 GB~89%
Q4_K_L
13.3 GB~90%
BF16
42.4 GB~100%
★ Recommended — best balance of quality and VRAM usage.
Learn more about quantization →
Can I run ERNIE 4.5 21B A3B PT on a Mac?: ERNIE 4.5 21B A3B PT requires at least 6.2 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
Can I run ERNIE 4.5 21B A3B PT locally?: Yes — ERNIE 4.5 21B A3B PT can run locally on consumer hardware. At Q4_K_M quantization it needs 13.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is ERNIE 4.5 21B A3B PT?: At Q4_K_M, ERNIE 4.5 21B A3B PT can reach ~338 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~50 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 13.0 × 0.65 = ~399 tok/s
Estimated speed at Q4_K_M (13.0 GB)
NVIDIA B200
~399 tok/s
NVIDIA GeForce RTX 4090
~50 tok/s
NVIDIA B300
~399 tok/s
AMD Instinct MI350X
~338 tok/s
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
Learn more about tok/s estimation →
What's the download size of ERNIE 4.5 21B A3B PT?: At Q4_K_M, the download is about 12.60 GB. The full-precision BF16 version is 42.00 GB. The smallest option (IQ2_XXS) is 5.78 GB.
Which GPUs can run ERNIE 4.5 21B A3B PT?: 26 consumer GPUs can run ERNIE 4.5 21B A3B PT at Q4_K_M (13.0 GB). Top options include AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090, AMD Radeon RX 6800. 8 GPUs have plenty of headroom for comfortable inference.
Which devices can run ERNIE 4.5 21B A3B PT?: 49 devices with unified memory can run ERNIE 4.5 21B A3B PT at Q4_K_M (13.0 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.