Alibaba (Qwen), open weights
Qwen3.5-9B
Qwen3.5-9B by Alibaba (Qwen) ranks 236th of 354 ranked models on the Noometry Index as of October 2026, with a score of 33.8. Its strongest category is knowledge, where it ranks 84th. API pricing starts at $0.10 per million input tokens and $0.15 per million output tokens, with a 262K-token context window.
Last verified
Specifications
- Noometry rank
- #236 of 354
- Index score
- 33.8
- Evidence
- Confirmed 10 results
- Provider
Alibaba (Qwen)- Released
- February 23, 2026
- Weights
- Open weights
- Reasoning
- Yes
- Context window
- 262K
- Max output
- 66K
- Input price
- $0.10 / M
- Output price
- $0.15 / M
- Blended price
- $0.11 / M
- Output speed
- Not measured
- Value
- #20 of 219
- Knowledge cutoff
- Unknown
- Input
- text, image
- Hugging Face
- Qwen/Qwen3.5-9B
Category scores
Each category score combines every public result we have in that category.
- Coding 35.9
- Agentic & Tool Use 14.5
- Reasoning 23.1
- Math 34.8
- Knowledge 46.0
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 35.9 | #217 | 1 |
| Agentic & Tool Use | 14.5 | #151 | 1 |
| Reasoning | 23.1 | #182 | 4 |
| Math | 34.8 | #192 | 2 |
| Knowledge | 46.0 | #84 | 1 |
Strengths and weaknesses
Categories where Qwen3.5-9B places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
Weakest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Agentic & Tool Use | 14.5 | −15.9 | #151 of 154, top 99% |
| Coding | 35.9 | −2.9 | #217 of 340, top 64% |
Closest competitors
The models ranked just above and below Qwen3.5-9B. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Qwen3 Coder Next | #232 | 34.3 | $0.29 | — | Compare |
| Devstral Small 2505 | #233 | 34.3 | $0.15 | 88 | Compare |
| Qwen1.5-110B | #234 | 34.2 | — | — | Compare |
| o1-mini | #235 | 34.0 | — | — | Compare |
| Codellama 70b Instruct | #237 | 33.7 | — | — | Compare |
| Qwen3 8B | #238 | 33.7 | $0.31 | — | Compare |
| Grok-2 (Dec 2024) | #239 | 33.7 | — | — | Compare |
| GPT-4.1 mini | #240 | 33.6 | $0.70 | 86 | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| SciCode | 27.5% | #108 of 121, top 90% | Epoch AI |
Agentic & Tool Use
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Terminal-Bench | 9.2% | #40 of 41, top 98% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| CritPt | 0.3% | #99 of 134, top 74% | Epoch AI | ||
| Chess Puzzles | 2% | Epoch AI | 2026-08-27 | ||
| Chess Puzzles | 12% | #81 of 129, top 63% | none | Epoch AI | 2026-08-27 |
| DTBench | 71.2% | #91 of 151, top 61% | Epoch AI | ||
| LMCA | 24.5% | #86 of 125, top 69% | Epoch AI | ||
| Epoch Capabilities Index | 139.46 | #111 of 213, top 53% | Epoch AI | 2026-02-24 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| MathArena Final-Answer Competitions | 48.5% | #27 of 29, top 94% | MathArena | ||
| OTIS Mock AIME 2024-2025 | 42.8% | Epoch AI | 2026-08-27 | ||
| OTIS Mock AIME 2024-2025 | 61.7% | #102 of 173, top 59% | none | Epoch AI | 2026-08-27 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | 79% | #83 of 186, top 45% | Epoch AI | 2026-08-27 | |
| GPQA Diamond | 78.9% | none | Epoch AI | 2026-08-27 |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| deepinfra | $0.10 | $0.15 | — | 2026-10-10 |
| openrouter | $0.10 | $0.15 | — | 2026-10-10 |
| together | $0.17 | $0.25 | — | 2026-10-10 |
Compare Qwen3.5-9B
- Qwen3.5-9B vs Qwen3.5 397B-A17B
- Qwen3.5-9B vs o1-mini
- Qwen3.5-9B vs Codellama 70b Instruct
- Qwen3.5-9B vs Qwen1.5-110B
- Qwen3.5-9B vs Qwen3 8B
- Qwen3.5-9B vs Devstral Small 2505
- Qwen3.5-9B vs Grok-2 (Dec 2024)
- Qwen3.5-9B vs GPT-6 Astra
- Qwen3.5-9B vs Claude Fable 5.1
- Qwen3.5-9B vs Gemini 3.8 Flash
- Qwen3.5-9B vs Kimi K3
- Qwen3.5-9B vs Grok 4.6
- Qwen3.5-9B vs GLM-5.3
- Qwen3.5-9B vs Muse Spark 1.3
Other Alibaba (Qwen) models
- Qwen3.8 Max56.8
- Qwen3.7 Max51.5
- Qwen3.6 Max Preview51.5
- Qwen3.6 Plus47.5
- Qwen3.5 397B-A17B46.0
- Qwen3.8 27B46.0
- Qwen3.5 Max Preview45.3
- Qwen3.7 Plus45.3
Frequently asked questions
How good is Qwen3.5-9B?
Qwen3.5-9B by Alibaba (Qwen) ranks 236th of 354 ranked models on the Noometry Index as of October 2026, with a score of 33.8. Its strongest category is knowledge, where it ranks 84th. API pricing starts at $0.10 per million input tokens and $0.15 per million output tokens, with a 262K-token context window.
How much does Qwen3.5-9B cost?
Qwen3.5-9B costs $0.10 per million input tokens and $0.15 per million output tokens on deepinfra.
What is Qwen3.5-9B's context window?
Qwen3.5-9B accepts up to 262K tokens of input and can write up to 66K tokens in one response.
Is Qwen3.5-9B open source?
Yes. Qwen3.5-9B's weights are downloadable from Hugging Face (Qwen/Qwen3.5-9B); check the license for commercial terms.
What are Qwen3.5-9B's strengths and weaknesses?
Relative to other ranked models, Qwen3.5-9B places best in knowledge, reasoning and lowest in agentic & tool use, coding.
What is Qwen3.5-9B best at?
Its best category is knowledge, where it ranks 84th on Noometry.