Qwen3.8-27B: Benchmarks, Pricing & Context Window
Qwen3.8-27B is a language model from Qwen, released in August 2026, with multimodal input, a 262K-token context window, and pricing from $0.400/M input, $0.040/M cached input, $3.00/M output.
Qwen3.8-27B is a dense 27.78-billion-parameter multimodal foundation model for coding, professional work, research, and long-horizon agents. It natively understands text, images, and video, supports configurable thinking effort and
Qwen3.8-27B benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for Qwen3.8-27B across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How Qwen3.8-27B holds up as conversations get longer.
Quality Tracker
Qwen3.8-27B Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Wed Sep 09 2026Qwen3.8-27B pricing
Providers
Qwen3.8-27B starts at $0.400 per million input tokens and $3.00 per million output tokens via DeepInfra. Reused prompt prefixes cost $0.0400 per million cached input tokens.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 DeepInfra | $0.400 | $0.0400 | $3.00 | 262.1K/262.1K | — | — | / |
 FriendliAI | — | — | — | 262.1K/131.1K | — | — | / |
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.
Qwen3.8-27B model size
Qwen3.8-27B has 27.8 billion parameters. See how it compares to other models in the same parameter range.
Qwen3.8-27B context window
Input and output token limits for Qwen3.8-27B, plus how it ranks on long-context understanding.
Try now
huggleMake it with
Qwen3.8-27B.
Qwen3.8-27B latency
Qwen3.8-27B time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
Qwen3.8-27B examples
Recent arena outputs from Qwen3.8-27B, picked from the highest-ranked matchups.
Qwen3.8-27B license
Qwen3.8-27B is released under the Apache 2.0 license, which permits commercial use, has 27.8B parameters.
- License
- Apache 2.0
- Commercial use allowed
- Parameters
- 27.8B
Apache License 2.0 - allows commercial use
Qwen3.8-27B resources
Official sources for Qwen3.8-27B: official playground, official launch post, source repository.
Qwen3.8-27B vs other models
The most-compared alternatives to Qwen3.8-27B are Sakana Namazu, Seed 2.0 Pro, MiMo-V2.5-Pro. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like Qwen3.8-27B
Models ranked just above and below Qwen3.8-27B by LLM Stats score.
FAQ
Common questions about Qwen3.8-27B.