Qwen 3.5 (9B)
Alibaba's Qwen 3.5 at 9B parameters. Multimodal (text + images), native tool calling, 256K context. Strong reasoning + tool use for devices with a 12GB+ inference budget — the small-tier sweet spot for tool-heavy work.
At a glance
- Parameters
- 9B
- Quantization
- Q4_K_M
- Context window
- 262K tokens
- Approx. size
- 6.6 GB
- Engines
- llama.cpp, MLX, Ollama
- License
- Apache-2.0 (open)
- Version
- v1.1.0 , released 2026-04-26
- Family
- qwen
Tuning
Defaults a gezel client applies out of the box, from real evaluation runs against this exact quantization.
- sampling.temperature
0.7- sampling.topP
0.8- sampling.topK
20- sampling.minP
0- sampling.maxTokens
16384- sampling.repetitionPenalty
1.1- sampling.repetitionContext
20- reasoning.thinkingBudget
2048- reasoning.enableThinking
true- profiles
- thinking-general, thinking-coding, thinking-precise, instruct, creative
Model behaviors
Client-side behavior modules the catalog enables for this model (reasoning-tag handling, fabrication detection, prompt shaping).
- reasoning.strip-think-tags
- prompt.private-reasoning-guidance
- prompt.tool-cookbook-condensed
- prompt.meester-build-prelude
- fabrication.detect-past-tense-no-tools
- fabrication.detect-claim-without-tool
- mcp.compact-tool-schemas
- mcp.relax-required-fields
- mcp.default-missing-fields
- turn.ollama-num-predict-bumped
- turn.preamble-folding
- turn.ramble-detection
- tools.mlx-grammar
- provider.merge-system-messages
- validate.inline-js-parses
- prompt.prefer-writefile-edits
- prompt.derive-by-execution
Sources
- llama.cpp (GGUF)
- lmstudio-community/Qwen3.5-9B-GGUF ·
Qwen3.5-9B-Q4_K_M.gguf· Q4_K_M - MLX
- mlx-community/Qwen3.5-9B-4bit
- Ollama
qwen3.5:9b- Upstream
- https://ollama.com/library/qwen3.5
- Manifest
- View on GitHub
Tags
- alibaba
- multimodal
- vision
- tools