Qwen3.5-397B-A17B: Benchmarks, Pricing & Context Window
Qwen3.5-397B-A17B is a language model from Qwen, released in February 2026, with multimodal input, a 262K-token context window, and pricing from $0.450/M input, $0.220/M cached input, $3.00/M output.
Qwen3.5-397B-A17B is Qwen's flagship Mixture-of-Experts model with 397 billion total parameters and 17 billion activated parameters. It delivers state-of-the-art performance across knowledge, reasoning, coding, mathematics, multilingual
Qwen3.5-397B-A17B benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for Qwen3.5-397B-A17B across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How Qwen3.5-397B-A17B holds up as conversations get longer.
Quality Tracker
Qwen3.5-397B-A17B Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Wed Oct 07 2026Qwen3.5-397B-A17B pricing
Providers
Qwen3.5-397B-A17B starts at $0.450 per million input tokens and $3.00 per million output tokens via DeepInfra. Reused prompt prefixes cost $0.220 per million cached input tokens. See all 2 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 DeepInfra | $0.450 | $0.220 | $3.00 | 262.1K/262.1K | — | — | / |
 Novita | $0.600 | — | $3.60 | 262.1K/64.0K | — | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
Qwen3.5-397B-A17B model size
Qwen3.5-397B-A17B has 397 billion parameters. See how it compares to other models in the same parameter range.
Qwen3.5-397B-A17B context window
Input and output token limits for Qwen3.5-397B-A17B, plus how it ranks on long-context understanding.
Try now
huggleMake it with
Qwen3.5-397B-A17B.
Qwen3.5-397B-A17B latency
Qwen3.5-397B-A17B time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
Qwen3.5-397B-A17B examples
Recent arena outputs from Qwen3.5-397B-A17B, picked from the highest-ranked matchups.
Qwen3.5-397B-A17B license
Qwen3.5-397B-A17B is released under the Apache 2.0 license, which permits commercial use, has 397.0B parameters.
- License
- Apache 2.0
- Commercial use allowed
- Parameters
- 397.0B
Apache License 2.0 - allows commercial use
Qwen3.5-397B-A17B resources
Official sources for Qwen3.5-397B-A17B: provider documentation, official playground, source repository, model weights.
Qwen3.5-397B-A17B vs other models
The most-compared alternatives to Qwen3.5-397B-A17B are Seed 2.0 Pro, Grok-4 Heavy, GPT-5.1 Thinking. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like Qwen3.5-397B-A17B
Models ranked just above and below Qwen3.5-397B-A17B by LLM Stats score.
FAQ
Common questions about Qwen3.5-397B-A17B.