Qwen 2.5 3B
Alibaba · dense · 3.09B parameters · 32,768 context
Parameters
3.09B
Context Window
32K tokens
Architecture
Dense
Best GPU
RTX 4070 Ti
Cheapest API
$0.10/M
Quality Score
58/100
Intelligence Brief
Qwen 2.5 3B is a 3.09B parameter DENSE model from Alibaba, featuring Grouped Query Attention (GQA) with 36 layers and 2,048 hidden dimensions. With a 32,768 token context window, it supports tools, structured output, code, math, multilingual. On standardized benchmarks, it achieves MMLU 65, HumanEval 35, GSM8K 70. The most cost-effective API deployment is via alibaba at $0.10/M output tokens. For self-hosted inference, RTX 4070 Ti delivers optimal throughput at $237/month.
Provider pricing
1 provider · canonical: alibaba| Provider | Input $/M | Output $/M ▲ | Notes |
|---|---|---|---|
| alibabacanonical | $0.100 | $0.100 | cheapest input · cheapest output |
Prices update via the nightly pricing cron + admin approvals at /admin/ingest-queue. The leaderboard's Input/Output cells show the canonical rate above; this table shows the full spread.
Recent changes
Loading…
Related models
5 suggestions- Qwen 2.5 14BQwen 2.5 · 14.8B$0.300/M out
- Qwen 2.5 32BQwen 2.5 · 32.5B$0.800/M out
- Qwen 2.5 Coder 32BQwen 2.5 · 32.5Bfree/M out
- Qwen 2.5 0.5BQwen 2.5 · 0.5B—
- Qwen 2.5 1.5BQwen 2.5 · 1.5B—
Picks: same family first, then same vendor within ±2× params, then top tag-overlap matches. Price shown is the cheapest Output $/M across providers — the row's page shows the canonical anchor.
Architecture Details
Memory Requirements
BF16 Weights
6.2 GB
FP8 Weights
3.1 GB
INT4 Weights
1.5 GB
GPU Compatibility Matrix
Qwen 2.5 3B is compatible with 100% of GPU configurations across 41 GPUs at 3 precision levels.
GPU Recommendations
BF16 · 1 GPU · vllm
100/100
score
Throughput
440.4 tok/s
Latency (ITL)
2.3ms
Est. TTFT
0ms
Cost/Month
$237
Cost/M Tokens
$0.21
BF16 · 1 GPU · vllm
100/100
score
Throughput
664.0 tok/s
Latency (ITL)
1.5ms
Est. TTFT
0ms
Cost/Month
$133
Cost/M Tokens
$0.08
BF16 · 1 GPU · vllm
100/100
score
Throughput
440.4 tok/s
Latency (ITL)
2.3ms
Est. TTFT
0ms
Cost/Month
$209
Cost/M Tokens
$0.18
Deployment Options
API Deployment
alibaba
$0.10/M
output tokens
Single GPU
RTX 4070 Ti
$237/mo
Min VRAM: 3 GB
Multi-GPU
RTX 4070 Ti
440.4 tok/s
Best available config
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| alibaba | $0.10 | $0.10 | Cheapest |
Cost Analysis
| Provider | Input $/M | Output $/M | ~Monthly Cost |
|---|---|---|---|
| alibabaBest Value | $0.10 | $0.10 | $1 |
Cost per 1,000 Requests
Short (500 tok)
$0.07
via alibaba
Medium (2K tok)
$0.28
via alibaba
Long (8K tok)
$1.00
via alibaba
Performance Estimates
Throughput by GPU
VRAM Breakdown (RTX 4070 Ti, BF16)
Precision Impact
bf16
6.2 GB
weights/GPU
~440.4 tok/s
fp8
3.1 GB
weights/GPU
int4
1.5 GB
weights/GPU
Quality Benchmarks
Capabilities
Features
Supported Frameworks
Supported Precisions
Where to Deploy Qwen 2.5 3B
Self-Hosted Infrastructure
Similar Models
Qwen 2.5 1.5B
1.5B params · dense
Quality: 50
Qwen 2.5 7B
7.6B params · dense
Quality: 70
from $0.00/M
Qwen 2.5 Coder 3B
3.1B params · dense
Quality: 50
StarCoder2 3B
3.03B params · dense
Quality: 29
from $0.10/M
OpenELM 3B
3B params · dense
Quality: 50
Frequently Asked Questions
How much VRAM does Qwen 2.5 3B need for inference?
Qwen 2.5 3B requires approximately 6.2 GB of VRAM at BF16 precision, 3.1 GB at FP8, or 1.5 GB at INT4 quantization. Additional VRAM is needed for KV-cache (36864 bytes per token) and activations (~0.50 GB).
What is the best GPU for Qwen 2.5 3B?
The top recommended GPU for Qwen 2.5 3B is the RTX 4070 Ti using BF16 precision. It achieves approximately 440.4 tokens/sec at an estimated cost of $237/month ($0.21/M tokens). Score: 100/100.
How much does Qwen 2.5 3B inference cost?
Qwen 2.5 3B API inference starts from $0.10/M input tokens and $0.10/M output tokens. Self-hosted inference costs depend on your GPU configuration — use our ROI calculator for a detailed breakdown.