Skip to content
Updated minutes ago
Alibaba

Qwen 3 8B

Alibaba · dense · 8.2B parameters · 131,072 context

Quality
67.0

Architecture Details

TypeDENSE
Total Parameters8.2B
Active Parameters8.2B
Layers36
Hidden Dimension4,096
Attention Heads32
KV Heads8
Head Dimension128
Vocab Size151,936

Memory Requirements

BF16 Weights

16.4 GB

FP8 Weights

8.2 GB

INT4 Weights

4.1 GB

KV-Cache per Token147456 bytes
Activation Estimate1.00 GB

Fits on (single-node)

B200 SXM BF16B100 SXM BF16GB200 NVL72 (per GPU) BF16GB300 NVL72 (per GPU) BF16H200 SXM BF16H100 SXM BF16H100 PCIe BF16H100 NVL BF16

GPU Recommendations

A30optimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

307.2 tok/s

Cost/Month

$332

Cost/M Tokens

$0.41

Use this config →
RTX 4090optimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

331.9 tok/s

Cost/Month

$370

Cost/M Tokens

$0.42

Use this config →
RTX 3090optimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

308.2 tok/s

Cost/Month

$180

Cost/M Tokens

$0.22

Use this config →

API Pricing Comparison

ProviderInput $/MOutput $/MBadges
together$0.20$0.20
Cheapest
fireworks$0.20$0.20

Quality Benchmarks

MMLU
72.0
HumanEval
42.0
GSM8K
78.0
MT-Bench
77.0

Capabilities

Features

Tool Use Vision Code Math Reasoning Multilingual Structured Output

Supported Frameworks

vllmsglangtgitensorrt-llmollama

Supported Precisions

BF16 (default)FP8INT4

Similar Models