About
Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per...
Related Models
Pricing
Input
$0.11
per 1M tokens
Output
$0.80
per 1M tokens
Blended
$0.28
per 1M tokens
Cheaper than 63% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.28
Monthly
$8.47
vs. Similar Models
DeepSeek V3.2 Exp (Non-reasoning)Q:+0.1
$0.31+10%
DeepSeek: DeepSeek V3.1 TerminusQ:+0.2
$0.44+56%
DeepSeek V3.1 (Non-reasoning)Q:-0.2
$0.83+195%
Nova 2.0 Omni (medium)Q:-0.3
$0.85+201%
Performance
127
tokens/sec
Faster than 66% of models
1.05
seconds
Faster than 53% of models
1.05
seconds
Faster than 67% of models
Market Median
94 tok/s
36% faster
Median TTFT
1.10s
5% faster
Throughput/Dollar
450
tok/s per $/1M
Speed Comparison
Nova 2.0 Pro Preview (medium)
127 tok/s-0%
Qwen3 VL 30B A3B (Reasoning)
124 tok/s-2%
Grok 4.3 (low)
130 tok/s+3%
Context Window
262K
tokens
Larger than 62% of models
Max Output
262K
tokens
100% of context
Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
73.7%
HLE
9.3%
LiveCodeBenchNot evaluated
SciCode
32.3%
TerminalBench Hard
18.2%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
35.2%
Long Context Recall
40.0%
Tau2
79.5%
Market AverageTop Score