standardApache 2.09.7B paramsOpen weights

Qwen3.5 9B

9.7B parameter open-weight model. The mid-small model of the Qwen3.5 generation — 9.7B parameters, 256K context, image input, Apache 2.0. Carries the same hybrid attention stack as the 27B checkpoints (a full-attention layer every fourth block), so no architectureNotes are published for this record. The Ollama library Q4_K_M is 6.6 GB, Q8_0 is 11 GB and BF16 is 19 GB; with the 1.5 GB runtime overhead this site grades against, Q4 asks 8.1 GB, so a 10 GB card is a tight fit and 12 GB is comfortable. Worth knowing why this generation is still current at this size: neither Qwen3.6 nor Qwen3.8 released a checkpoint below 27B, so Qwen3.5 remains the newest Qwen line under that mark. No throughput measured on our rigs yet. Figures verified against the Hugging Face model card, its config.json, and the Ollama library on 2026-08-19.

Alibaba · Qwen 3.5

Editorial review

Reviewed byOpenSourcesAI EditorialLast updatedJune 2026SourcesHuggingFace model card (Qwen/Qwen3.5-9B), official docs, OpenSourcesAI editorial review.

VRAM figures are empirical estimates. Actual usage varies by runtime, context length, and system configuration. Verify on your specific hardware before production use.

Will Qwen3.5 9B run on your machine?

Qwen3.5 9B is 9.7B parameters and needs 8.1 GB of VRAM at Q4_K_M — 6.6 GB of weights plus 1.5 GB of runtime overhead for the inference server itself.

VRAM by quantization

QuantizationWeightsNeeds (with overhead)Quality
Q4_K_M6.6 GB8.1 GBgood
Q8_011 GB12.5 GBhigh
FP1619 GB20.5 GBreference

Fit on common hardware at Q4_K_M

HardwareMemory the model can useSystem RAMVerdict
CPU OnlyNone (CPU only)16 GBCPU offload
RTX 4060 Laptop8 GB16 GBCPU offload
RTX 3060 (12GB)12 GB32 GBComfortable
RTX 4060 Ti (16GB)16 GB32 GBComfortable
RTX 309024 GB64 GBComfortable

Comfortable means VRAM clears the requirement by 2 GB or more. Tight means it covers the requirement with no margin. CPU offload means the model does not fit in VRAM but system RAM is at least 1.6× the weights, so it will run at reduced speed — expect roughly 1–5 tokens per second. Figures are weights plus a fixed runtime overhead and exclude KV-cache growth, which scales with context length.

Check Qwen3.5 9B against your exact hardwareHow these verdicts are calculated

Need more hardware for Qwen3.5 9B? Open the PC Builder for the 13B / 14B tier →

Related models

Qwen3.8 9B Distill9.7B parameter model — shared focus: reasoning, math, coding.Qwythos-9B (Claude-Mythos-5 1M)9B parameter model — shared focus: reasoning, coding, tool-use.DeepSeek-R1 8B (0528 Qwen3)8B parameter model — shared focus: reasoning, coding, chat.Qwen2.5 14BBuilders hosting local AI on a single RTX 4080 (16 GB) or RTX 4090 (24 GB) who want near-frontier multilingual reasoning, mathematical accuracy, and agentic coding performance without multi-GPU infrastructure. The sweet spot between consumer accessibility and frontier-class open-weight capability for 2026 local deployment.

Ready to run this model locally?

Find a compatible interface in our Local AI Tools directory →