Updated minutes ago
Qwen 2.5 Coder 7B
Alibaba · dense · 7.6B parameters · 131,072 context
Quality50.0
Architecture Details
TypeDENSE
Total Parameters7.6B
Active Parameters7.6B
Layers28
Hidden Dimension3,584
Attention Heads28
KV Heads4
Head Dimension128
Vocab Size152,064
Memory Requirements
BF16 Weights
15.2 GB
FP8 Weights
7.6 GB
INT4 Weights
3.8 GB
KV-Cache per Token57344 bytes
Activation Estimate0.80 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
A10Goptimal
BF16 · 1 GPU · vllm
100/100
score
Throughput
213.1 tok/s
Cost/Month
$285
Cost/M Tokens
$0.51
A30optimal
BF16 · 1 GPU · vllm
100/100
score
Throughput
331.4 tok/s
Cost/Month
$332
Cost/M Tokens
$0.38
RTX 4090optimal
BF16 · 1 GPU · vllm
100/100
score
Throughput
358.1 tok/s
Cost/Month
$370
Cost/M Tokens
$0.39
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| together | $0.20 | $0.20 | Cheapest |
Capabilities
Features
✗ Tool Use✗ Vision✓ Code✓ Math✗ Reasoning✗ Multilingual✓ Structured Output
Supported Frameworks
vllmsglangtgitensorrt-llmollama
Supported Precisions
BF16 (default)FP8INT4