FitMyLLM
Baidu/Dense

BERNIE-4.5-300B-A47B-Paddle

> Note: "-Paddle" models use PaddlePaddle weights, while "-PT" models use Transformer-style PyTorch weights.

chat
300.5B
Parameters
128K
Context length
13
Benchmarks
17
Quantizations
1K
HF downloads
Architecture
Dense
Released
2025-06-28
Layers
54
KV Heads
8
Head Dim
128
Family
ernie

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
IQ2_XXS2.38
92.4 GB
89.9 + 2.5 KV
low
IQ2_M2.93
113.1 GB
110.5 + 2.5 KV
low
Q2_K3.16
121.7 GB
119.2 + 2.5 KV
low
IQ3_XXS3.25
125.1 GB
122.6 + 2.5 KV
low
IQ3_XS3.5
134.5 GB
132.0 + 2.5 KV
low
Q3_K_S3.64
139.7 GB
137.2 + 2.5 KV
low
IQ3_M3.76
144.3 GB
141.7 + 2.5 KV
low
Q3_K_M4
153.3 GB
150.7 + 2.5 KV
low
Q3_K_L4.3
164.5 GB
162.0 + 2.5 KV
moderate
IQ4_XS4.46
170.5 GB
168.0 + 2.5 KV
moderate
Q4_K_S4.67
178.4 GB
175.9 + 2.5 KV
moderate
Q4_K_M4.89
186.7 GB
184.2 + 2.5 KV
good
Q5_K_S5.57
212.2 GB
209.7 + 2.5 KV
good
Q5_K_M5.7
217.1 GB
214.6 + 2.5 KV
good
Q6_K6.56
249.4 GB
246.9 + 2.5 KV
excellent
Q8_08.5
322.3 GB
319.8 + 2.5 KV
lossless
FP1616
604.0 GB
601.5 + 2.5 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Too big for a single GPU — plan a multi-GPU deployment
Even the lightest quant needs ~92 GB. Size GPUs, replicas, TCO and scaling for a production setup. Open in Enterprise →
READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN ERNIE-4.5-300B-A47B-PADDLE NOW

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (13)

MATH-50093.1
MMLU-PRO77.6
GPQA Diamond74.0
LiveCodeBench46.7
AIME41.3
AA Math41.3
IFBench39.1
SciCode31.5
AA Intelligence15.0
AA Coding14.5
Terminal-Bench6.1
HLE3.5
AA Long Context2.3

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run ernie:301b-q4_K_M

Tag may need adjustment — check ollama.com/library/ernie for available tags.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

AMD Instinct MI300X
192 GB VRAM • 5300 GB/s
AMD
$15000
Apple M2 Ultra (192GB)
192 GB VRAM • 800 GB/s
APPLE
$5499
Apple M3 Ultra (192GB)
192 GB VRAM • 800 GB/s
APPLE
$6999
Apple M4 Ultra (192GB)
192 GB VRAM • 1092 GB/s
APPLE
$7499
AMD Radeon Instinct MI300A
192 GB VRAM • 10300 GB/s
AMD
$12000
AMD Radeon Instinct MI300X
192 GB VRAM • 10300 GB/s
AMD
$15000
AMD Radeon Instinct MI308X
192 GB VRAM • 10300 GB/s
AMD
$12000
Apple M5 Ultra (192GB)
192 GB VRAM • 1228 GB/s
APPLE
AMD Radeon Instinct MI325X
288 GB VRAM • 10300 GB/s
AMD
$20000
AMD Radeon Instinct MI350X
288 GB VRAM • 8190 GB/s
AMD
$25000
AMD Radeon Instinct MI355X
288 GB VRAM • 8190 GB/s
AMD
$30000
Apple M4 Ultra (384GB)
384 GB VRAM • 1092 GB/s
APPLE
$9999
Apple M5 Ultra (384GB)
384 GB VRAM • 1228 GB/s
APPLE

Find the best GPU for ERNIE-4.5-300B-A47B-Paddle

Build Hardware for ERNIE-4.5-300B-A47B-Paddle
▸ SPEC SHEET

ERNIE-4.5-300B-A47B-Paddle300.5B Dense.

▸ SPECIFICATIONS
PARAMETERS
300.5B
ARCHITECTURE
Dense Transformer
CONTEXT LENGTH
128K tokens
CAPABILITIES
chat
RELEASE DATE
2025-06-28
PROVIDER
Baidu
FAMILY
ernie
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ2_XXS2.3889.9 GB65%
IQ2_M2.93110.5 GB75%
Q2_K3.16119.2 GB78%
IQ3_XXS3.25122.6 GB82%
IQ3_XS3.5132.0 GB84%
Q3_K_S3.64137.2 GB85%
IQ3_M3.76141.7 GB86%
Q3_K_M4150.7 GB88%
Q3_K_L4.3162.0 GB90%
IQ4_XS4.46168.0 GB92%
Q4_K_S4.67175.9 GB93%
Q4_K_M4.89184.2 GB94%
Q5_K_S5.57209.7 GB96%
Q5_K_M5.7214.6 GB96%
Q6_K6.56246.9 GB97%
Q8_08.5319.8 GB100%
FP1616601.5 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO77.6
GPQA Diamond74.0
aa_ifbench39.1
aa_terminal_bench6.1
aa_scicode31.5
aa_lcr2.3
LiveCodeBench46.7
AIME41.3
MATH-50093.1
HLE3.5
AA Intelligence15.0
AA Coding14.5
AA Math41.3
§ 02RUN COMMAND

Run ERNIE-4.5-300B-A47B-Paddle locally with Ollama — needs 184.2 GB VRAM at Q4_K_M:

$ollama run ernie:300b
§ 03COMPATIBLE GPUs
13 @ Q4_K_M