Qwen3.5 397B A17B
Q4 GGUF download: unsloth/Qwen3.5-397B-A17B-GGUF - 0.01 GB on disk, read 2026-09-27. VRAM figures above are estimates; this is the actual file size.
Intelligence benchmarks
Artificial Analysis indexes - compared with the best open and proprietary models
Intelligence
21.4
AA Index
Intelligence Index - Qwen3.5 397B A17B vs. the field
Best open-weight models (you can run locally) and leading proprietary models for context.
Benchmark data from Artificial Analysis · updated 2026-10-19.
Hosted API cost & speed
What it costs to rent this model instead of running it locally
| Metric | Value |
|---|---|
| Input (per 1M tokens) | $0.60 |
| Output (per 1M tokens) | $3.60 |
| Median speed (hosted API) | 57 tok/s |
Median across hosted providers, measured by Artificial Analysis - updated 2026-10-19. Local llama.cpp speed depends on your hardware — see the hardware pages for estimates.
Arena rating (LMArena)
Human preference wins from head-to-head chat battles
| Metric | Value |
|---|---|
| Overall arena rating | 1437.7 |
| Rank (overall) | #81 |
| Battle votes | 93,567 |
Overall rating published by LMArena on 2026-10-15 - a pairwise human-preference Elo across live chat battles; higher is better. Fetched 2026-10-19. Open-weights models compete head-to-head with proprietary ones here — no benchmark prompts, just what people pick.
Standard benchmarks
Performance across standard evaluations
| Benchmark | Score |
|---|---|
| MMLU-Pro | 87.8 |
| GPQA Diamond | 86.1 |
| AIME | 93.3 |
| SWEBENCH | 76.4 |
| HLE | 19.8 |
| IFBench | 51.6 |
| Terminal-Bench | 35.6 |
Source: Artificial Analysis component evals - updated 2026-10-19. Additional scores from Hugging Face OpenEvals official leaderboards - updated 2026-10-19.
Will it run on your hardware?
Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest
Need an exact figure for your context length? Use the VRAM calculator.
Run it locally
Copy-paste - running in under a minute
vllm serve Qwen/Qwen3.5-397B-A17BNew to this? Start with Ollama · serve to many users with vLLM.
Deep dive
Notes, sources, and the full write-up
Qwen3.5 397B A17B
Qwen3.5 397B A17B is a 403.4B-parameter apache-2.0 model from Alibaba. It scores 21.4 on the Artificial Analysis Intelligence Index (coding 41.3). At Q4_K_M it needs roughly 234 GB of VRAM, placing it in the 48gb+ hardware tier.
Specifications
| Spec | Value |
|---|---|
| Parameters | 403.4B |
| Context length | 262K tokens |
| License | apache-2.0 |
| Modalities | text, vision |
| Released | 2026-02-16 |
| Weights | Qwen/Qwen3.5-397B-A17B |
Benchmarks
Artificial Analysis Intelligence Index - Qwen3.5 397B A17B vs. leading closed models:
| Model | Intelligence | Coding | GPQA |
|---|---|---|---|
| Qwen3.5 397B A17B | 45 | 41.3 | 89.3 |
| GPT-5.5 (xhigh) | 60.2 | 59.1 | 93.5 |
| Claude Opus 4.8 (max) | 61.4 | 56.7 | 92 |
| Gemini 3.1 Pro Preview | 57.2 | 55.5 | 94.1 |
| Grok 4.3 (high) | 53.2 | 41 | 90.1 |
Source: Artificial Analysis (2026-06-04).
VRAM requirements
| Quant | VRAM | Runs on |
|---|---|---|
| Q4_K_M | ~234 GB | multi-GPU / datacenter |
| Q5_K_M | ~286 GB | multi-GPU / datacenter |
| Q8_0 | ~432 GB | multi-GPU / datacenter |
| FP16 | ~807 GB | multi-GPU / datacenter |
VRAM is estimated from parameter count; MoE models still need all weights resident.
How to run
vLLM:
vllm serve Qwen/Qwen3.5-397B-A17BPopularity
Qwen3.5 397B A17B has 1,083,473 downloads in the last month on HuggingFace and 1,501 likes.
Frequently asked
Quick answers to common questions
How much VRAM does Qwen3.5 397B A17B need?
Qwen3.5 397B A17B with 403.4B parameters needs approximately 234 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.
Is Qwen3.5 397B A17B better than other Qwen models?
Qwen3.5 397B A17B has 403.4B parameters with 262,144 context - a strong choice for general use.
What license is Qwen3.5 397B A17B under?
Qwen3.5 397B A17B is released under the apache-2.0 license, making it suitable for most commercial and personal projects.
What hardware runs Qwen3.5 397B A17B well?
With 403.4B parameters, Qwen3.5 397B A17B requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.
What is the best quantization for Qwen3.5 397B A17B?
Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~286 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.
How long can Qwen3.5 397B A17B's context window handle?
Qwen3.5 397B A17B supports a 262,144-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.
What models compete with Qwen3.5 397B A17B?
Qwen3.5 397B A17B competes with other 202B–605B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.
Nearby options
Similar models and compatible hardware by spec