About
Qwen2.5-Coder-7B-Instruct is a 7B parameter instruction-tuned language model optimized for code-related tasks such as code generation, reasoning, and bug fixing. Based on the Qwen2.5 architecture, it incorporates enhancements like RoPE,...
Related Models
Pricing
Input
$0.03
per 1M tokens
Output
$0.09
per 1M tokens
Blended
$0.04
per 1M tokens
Cheaper than 90% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.04
Monthly
$1.35
vs. Similar Models
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)Q:+0.1
$0.30+567%
Mistral LargeQ:-0.1
$3.00+6567%
Qwen3.5 0.8B (Non-reasoning)Q:-0.1
$0.02-56%
Claude 3 SonnetQ:+0.2
$6.00+13233%
Performance
Context Window
33K
tokens
Larger than 7% of models
Context Window Comparison
Mistral: Saba
33KSame
Mixtral 8x7B Instruct
33KSame
LFM2 8B A1B
33KSame
Benchmarks
MMLU-Pro
47.3%
GPQA Diamond
33.9%
HLE
4.8%
LiveCodeBench
12.6%
SciCode
14.8%
TerminalBench HardNot evaluated
MATH-500
66.0%
AIME
5.3%
AIME 2025Not evaluated
IFBenchNot evaluated
Long Context RecallNot evaluated
Tau2Not evaluated
Market AverageTop Score
Open Source
apache-2.07BGGUF / GPTQ / AWQ
Downloads
2.0M
Likes
743
VRAM (FP16)
8-16 GB
GPU
RTX 4070 / M2 Pro