Ctrl K
Back to models

Alibaba

Qwen3.5 9B

  • Released
  • Open weights
  • Apache 2.0
Compare
Intelligence Index

Reasoning setting

11.2#68 of 83measured models
Measured · Artificial Analysis v4.3.2
API price from
$0.10 in / $0.15 out / 1M tokOpenRouter · DeepInfra · BF16
Context window
262.14K tok

Performance profile

ReasoningRank #51 of 107
80.6%
GPQA Diamond
CodingRank #82 of 94
29.5%
SciCode
Cost efficiencyRank #37 of 67
52.7pts / $
Intelligence ÷ eval cost

Capabilities

Context window

Modalities

Measured setting

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
Qwen3.5 9B
Cost per task (USD) · log scale
Qwen3.5 9BReasoning
Intelligence Index
11.2
Cost / task
$0.2126
Evaluation time
8.7m

Pricing & access

API route
OpenRouter · DeepInfra · BF16API
Tier & contextStandard

USD per 1M tokens

Input
$0.10
Output
$0.15

Run locally

GGUF
Context
Estimated memory

~7.5 GiB

RAM + VRAM

Estimate assumptions

One text chat at the selected context, with a 16-bit cache and app allowance. Operating system and image processing are extra; optimized caches can use less.

16 GiB RAMwithout GPU offload

Memory budgetOne text chat
Weights
5.2 GiB
Context + state
0.3 GiB16-bit cache
App allowance
2 GiB
Disk space
~13 GBIncludes image file
Available software
Ollama (opens in a new tab)LM Studio (opens in a new tab)Jan (opens in a new tab)KoboldCpp (opens in a new tab)llama.cpp (opens in a new tab)
Get GGUF (opens in a new tab)
Set up in Ollama · Q4_K_M · 8K context

Save the following as Modelfile. The context matches the estimate above.

FROM hf.co/lmstudio-community/Qwen3.5-9B-GGUF:Q4_K_M
PARAMETER num_ctx 8192

Then run these commands in your terminal:

ollama create llmcurrent-local -f Modelfile
ollama run llmcurrent-local

Ollama import instructions (opens in a new tab) · For LM Studio or Jan, open the app’s model search, enter lmstudio-community/Qwen3.5-9B-GGUF, choose Q4_K_M, and set context to 8,192 tokens in the app’s model settings. Software links above open each app’s setup guide.

Images need extra memoryvLLM setup (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Qwen3.5 9B · Reasoning#77 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 74KAT Coder Pro V2Standard16.1%
  6. 75Trinity Large ThinkingReasoning15.8%
  7. 76Gemma 4 12BReasoning15.7%
  8. 77Qwen3.5 9BReasoning14.9%
  9. 78Qwen3.5 Omni PlusStandard14.9%
  10. 79K2 Horizon 3.7BReasoning13.9%
  11. 80Mistral Medium 3.5High13.8%

Peers use their best measured result on this benchmark.

Alternatives

Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.