Granite 4.0 H-Small 32B-A9BvsQwen 3.6 27B
Complete comparison between Granite 4.0 H-Small 32B-A9B (32B parameters, IBM) and Qwen 3.6 27B (27B, Alibaba). VRAM required by quantization, tokens/second measured on 4 reference GPUs, use-case verdict, license, installation commands. All figures are calculated from catalog data—no copy-pasting between pages.
In brief
| Feature | Granite 4.0 H-Small 32B-A9B | Qwen 3.6 27B |
|---|---|---|
| Settings | 32B | 27B |
| Family | Granite | Qwen |
| Author | IBM | Alibaba |
| Origin | US | CN |
| License | Apache 2.0 | Apache 2.0 |
| Context | 128,000 tokens | 262,144 tokens |
| Output | October 2025 | April 2026 |
Memory footprint
Approximate VRAM required for inference with an average context window. The winner (in green) is the model that uses less — smaller models have the edge.
| Quantization | Granite 4.0 H-Small 32B-A9B | Qwen 3.6 27B |
|---|---|---|
| Q4_K_M (lightweight) | 19 GB | 16 GB |
| Q5_K_M (balanced) | 23 GB | 19 GB |
| Q8 (near-lossless) | 35 GB | 29 GB |
| FP16 (max quality) | 64 GB | 54 GB |
| CPU-only RAM | 32 GB | 28 GB |
Estimated speed (tokens/second)
Estimates based on the best quantization that can fit on each GPU. Actual figures depend on the prompt, context, and engine (llama.cpp, vLLM, MLX). Methodology.
| Reference GPU | Granite 4.0 H-Small 32B-A9B | Qwen 3.6 27B |
|---|---|---|
| RTX 4090 (24 GB) | 75 tok/s · Q5_K_M | 32 tok/s · Q5_K_M |
| RTX 4080 (16 GB) | ✗ too heavy | 13 tok/s · Q4_K_M |
| RTX 3060 12GB (12 GB) | ✗ too heavy | ✗ too heavy |
| Apple M4 Pro (48 GB) (36 GB) | 30 tok/s · Q8 | 13 tok/s · Q8 |
Verdict by use case
For each common use case, we indicate which of the two is better suited based on its tags, size, and specialization.
Your private, free ChatGPT on your machine in 1 hour — LM Studio, Ollama, Open WebUI, your documents, no cloud.
- Lifetime online access
- PDF + files
- 30-day money-back guarantee
Strengths and weaknesses
Granite 4.0 H-Small 32B-A9B
Mamba-2 + MoE 32B/9B hybrid. ~70% less RAM in long contexts. Apache 2.0.
- 70% less RAM in long-context workloads
- Apache 2.0
- Enterprise-ready
- Recent llama.cpp support required
Installation
Qwen 3.6 27B
Dense multimodal 27B released April 22, 2026. 262k ctx (1M YaRN). SWE-bench Verified 77.2%.
- SWE-bench Verified 77.2%
- Native multimodal
- 262k context (1M YaRN)
- Apache 2.0
- 16+ GB in Q4
- Hybrid architecture (recent llama.cpp support required)
Installation
Frequently asked questions
Which model runs best on RTX 4090 (24 GB): Granite 4.0 H-Small 32B-A9B or Qwen 3.6 27B?
On an RTX 4090, Granite 4.0 H-Small 32B-A9B runs at Q5_K_M (~75 tok/s), Qwen 3.6 27B in Q5_K_M (~32 tok/s). In pure throughput terms, Granite 4.0 H-Small 32B-A9B wins. See the configurator to test your exact GPU.
Which uses the least VRAM?
In Q4_K_M, Qwen 3.6 27B fits in 16 GB versus 19 GB for its rival—a 3 GB gap, significant if you're targeting an RTX 3060 or an 8 GB 4060 Ti.
Granite 4.0 H-Small 32B-A9B or Qwen 3.6 27B for coding?
Qwen 3.6 27B is the best fit (code-specialized, top HumanEval scores). For an IDE copilot, pair it with Continue or Aider.
Can these models be used in commercial production?
Granite 4.0 H-Small 32B-A9B is under Apache 2.0 — free for commercial use. Qwen 3.6 27B is under Apache 2.0 — also free to use. For SaaS, prefer Apache 2.0 or MIT.
Which one should you choose in 2026?
Depends on your primary constraint. Smaller / faster: Qwen 3.6 27B (27B). More capable: Granite 4.0 H-Small 32B-A9B (32B). If you're unsure, run the configurator with your GPU and use case — it will decide based on both.
Alternatives to consider
If neither works for you, here are some neighboring models you could explore.