Scheduled for retirement on Oct 21, 2026. Still available at the prices below until then.
Context
262K
Max output
33K
Serving providers
3
Cheapest input
$0.10 /1M
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: Artificial AnalysisSpeed & latency
Output speed
66 t/s
tokens / second
Time to first token
1.90s
latency
Headline indices
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Automation Bench | 0.0 | default | — | |||||||||||||||||||||
| CritPt | 1% | reasoning: false | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| EvalComputeProxyFrozen Aug 18, 2026 | 2028.4 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| GDPval v2Frozen Aug 18, 2026 | 643.8 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| GPQA Diamond | 81% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Humanity's Last Exam | 15% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| IFBench | 67% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Long-Context Reasoning | 70% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| MMMU-Pro | 69% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience | -53.5 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience Accuracy | 0.2 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience Non Hallucination | 0.2 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| SciCode | 30% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Terminal Bench40 | 0.0 | default | — | |||||||||||||||||||||
| TerminalBench Hard | 24% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| TerminalBench v2.1 | 29% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| τ-bench Banking | 7% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| τ²-bench | 87% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
Artificial Analysis last published scores for this benchmark on Sep 9, 2026. Scores here are frozen at that date, and models released since are not on it.
Artificial Analysis last published scores for this benchmark on Aug 18, 2026. Scores here are frozen at that date, and models released since are not on it.
Pricing by serving provider
standard tier · per 1M tokens · click a provider to expandInput pricing runs from $0.10 to $0.17 per 1M tokens across 3 providers, so the dearest route costs 70% more than the cheapest for the same model.
| Serving provider | Input /1M | Output /1M | Endpoints | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfraCheapest | $0.10 | $0.15 | 1 | ||||||||||
Standarddeepinfra/Qwen/Qwen3.5-9B
| |||||||||||||
| OpenRouter | $0.10 | $0.15 | 1 | ||||||||||
Standardqwen/qwen3.5-9b
| |||||||||||||
| Together AI | $0.17 | $0.25 | 1 | ||||||||||
Standardtogether_ai/Qwen/Qwen3.5-9B
| |||||||||||||
| OpenRouter (batch) | $0.17 | $0.25 | 1 | ||||||||||
Batchqwen/qwen3.5-9b:batch
| |||||||||||||
Price history
listed seller prices, input and output per 1M tokens Input Output
Input down 41% since first tracked
Cost calculator
estimate your monthly spend on this model$32
estimated / month
Model IDs
copy the exact identifier for your platformqwen/qwen3.5-9bdeepinfra/Qwen/Qwen3.5-9Bqwen/qwen3.5-9b:batchtogether_ai/Qwen/Qwen3.5-9B
Frequently asked questions
Qwen3.5-9B pricing, context and availabilityIs Qwen3.5-9B still available?
Yes. Qwen3.5-9B is scheduled for retirement on 2026-10-21.
How much does Qwen3.5-9B cost?
Qwen3.5-9B costs $0.10 per 1M input tokens from DeepInfra, and $0.15 per 1M output tokens from DeepInfra. Across 3 serving providers, input prices range from $0.10 to $0.17 per 1M tokens.
What is the context window of Qwen3.5-9B?
Qwen3.5-9B accepts up to 262K tokens of context and can return up to 33K output tokens.
Which providers serve Qwen3.5-9B?
Qwen3.5-9B is available from 3 serving providers, each with its own pricing and model ID. DeepInfra is currently the cheapest.
What can Qwen3.5-9B do?
Qwen3.5-9B supports image input (vision), tool use, reasoning and structured output.
Other Alibaba models
compare pricing across the Alibaba lineupQwen3 Next 80B A3B Instruct$0.09/1M · 11 providersQwen3 32B$0.08/1M · 10 providersQwen3 Next 80B A3B Thinking$0.14/1M · 10 providersQwen3.8 Max$1.65/1M · 10 providersQwen3.6 35B A3B$0.05/1M · 9 providersQwen3 VL 235B A22B Instruct$0.20/1M · 9 providersQwen2.5 Coder 32B Instruct$0.05/1M · 8 providersQwen3 235B A22b Instruct 2507$0.09/1M · 8 providers
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI