Qwen3 Coder Next
by Alibaba (Qwen) · qwen/qwen3-coder-next · #111 cheapest of 422 paid models
Prices updated Sep 19, 2026, 6:07 PM UTC · refreshed hourly
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...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.12
per 1M tokens
Output
$0.80
per 1M tokens
Blended (3:1)
$0.29
3 input : 1 output
Cache read
$0.07
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Qwen3 Coder Next with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.29 | $0.013 | — |
| 50% | $0.271 | $0.0105 | −19% |
| 90% | $0.256 | $0.00854 | −35% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.0003 | $0.3 |
| Document summary | 8,000 | 1,000 | $0.00176 | $1.76 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0052 | $5.20 |
| Long-context analysis | 150,000 | 5,000 | $0.022 | $22.00 |
Performance
Intelligence Index
10.1
AA composite quality
Coding Index
36.2
Math Index
—
Agentic Index
3.6
Output speed
125 tok/s
median
Time to first token
0.84s
median
Time to first answer
0.84s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| GPQA Diamond | 73.7% |
| Humanity's Last Exam | 10.1% |
| SciCode | 36.2% |
| IFBench | 35.2% |
| AA-LCR (Long Context Reasoning) | 47.0% |
| Terminal-Bench Hard | 18.2% |
| Terminal-Bench 2.1 | 38.2% |
| τ²-Bench (Telecom) | 79.5% |
| τ³-Bench Banking | 5.4% |
Specs
Context window
262K
tokens
Max output
236K
tokens
Input modalities
text
Tokenizer
Qwen