Updated minutes ago
Gemma 2 9B
Google · dense · 9.2B parameters · 8,192 context
Quality68.0
Architecture Details
TypeDENSE
Total Parameters9.2B
Active Parameters9.2B
Layers42
Hidden Dimension3,584
Attention Heads16
KV Heads8
Head Dimension256
Vocab Size256,000
Memory Requirements
BF16 Weights
18.4 GB
FP8 Weights
9.2 GB
INT4 Weights
4.6 GB
KV-Cache per Token344064 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
RTX 5090optimal
BF16 · 1 GPU · vllm
95/100
score
Throughput
473.3 tok/s
Cost/Month
$845
Cost/M Tokens
$0.68
V100 32GBoptimal
BF16 · 1 GPU · vllm
95/100
score
Throughput
91.7 tok/s
Cost/Month
$180
Cost/M Tokens
$0.75
Instinct MI100optimal
BF16 · 1 GPU · vllm
95/100
score
Throughput
110.8 tok/s
Cost/Month
$380
Cost/M Tokens
$1.30
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| deepinfra | $0.10 | $0.10 | Cheapest |
| together | $0.20 | $0.20 |
Quality Benchmarks
MMLU71.3
HumanEval40.0
GSM8K76.0
MT-Bench78.0
Capabilities
Features
✗ Tool Use✗ Vision✓ Code✓ Math✗ Reasoning✓ Multilingual✓ Structured Output
Supported Frameworks
vllmsglangtgitensorrt-llmollama
Supported Precisions
BF16 (default)FP8INT4