Qwen3.5-9B on local hardware
Estimated comfort for repository coding with Aider on a MacBook Pro M4 Pro 48GB.
Qwen3.5-9BOpen source
Qwen · apache-2.0 · 7,483,178 downloads
VisionReasoning / thinking mode
- Parameters
- 9.7B exact
- Context window
- 256K tokens
- Attention
- Hybrid linear + full attention
- Layers · KV heads · head dim
- 32 · 4 · 256
- KV cache
- 32 KB/token · 1.0 GB at 32K
- Tool calling
- good (from chat template)
- Thinking
- good (from chat template)
- Vision
- good (from config)
- Commercial use
- allowedRetain the required license and attribution notices
- Import confidence
- medium · capabilities estimated
Weights by quantization
Q3_K_M
4.4 GB
gguf file size
Q4_K_M
5.3 GB
gguf file size
Q5_K_M
6.1 GB
gguf file size
Q6_K
6.9 GB
gguf file size
Q8_0
8.9 GB
gguf file size
F16 / BF16
17 GB
gguf file size
- • vision: good, inferred from config (low confidence).
- • Capability ratings are estimated from model size and name — this model hasn't been assessed by hand.
- • GGUF sizes come from community conversions matched by model name; equivalence and runtime support are not verified. MLX conversion availability is unverified.
Qwen3.5-9B — 6-bit
Acceptable for your workload
Speed
20–25 tok/s
Adequate
Memory left
27 GB
of 48 GB
First word in
≈45 s
Adequate
All the details
- Your setup
- MacBook Pro M4 Pro 48 GB
- Workload
- Repository coding with Aider
- Runtime → tool
- LM Studio → Aider
- Memory used
- 9.6 GB
- Context
- 32K · up to 128K practical
- Tool support
- Official
- Use-case fit
- Poorly suited
Why?
- Model fits with approximately 27 GB of memory headroom.
- LM Studio supports hardware acceleration on this machine (Metal (Apple GPU)).
- Aider can connect to LM Studio (officially supported).
- Generation (20–25 tok/s) is adequate but not ideal for interactive coding.
- Ingesting a prompt of 14K tokens takes ≈45 s, and each step ≈55 s — noticeable waits.
- 32K window; generation drops to 19–25 tok/s when it is full.
- Weak coding capability for interactive coding.
Why “Acceptable” rather than “Comfortable”?
- Workload suitability: Weak coding capability for interactive coding.
- Prompt processing: Ingesting the first prompt (14K tokens) takes ≈45 s; each of the ~4 follow-up turns takes ≈55 s (≈5 min per task).
- Generation: Expected 20–25 tok/s at ~16K tokens of context. Coding with repository context is comfortable from about 28 tok/s.
- Context: 32K window; generation drops to 19–25 tok/s when it is full.
How sure are we?
Medium confidenceEstimated from bandwidth and compute, calibrated against 2 verified benchmarks of other models on this same chip (measured with a different engine, so calibration is partial).
- Calibrated with verified benchmarks of other models on this chip.
- Hybrid/sliding attention performance varies more between runtimes.
Medium confidence