Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding
Capability badges appear only where the provider catalog explicitly lists support. A blank cell means the source is silent, not that the feature is absent.
Listed rates
Price across providers
Input price per million tokens as published by each provider. Bars are drawn from listed rates only — no traffic weighting, since the catalog observes no requests.
Lowest input$0.1/M
Across 5 priced providers
Median input$0.2/M
Midpoint of listed rates
Highest input$0.201/M
2.0× the lowest listed rate
Output range$0.2 – $0.5
Per million output tokens
Vultr$0.1/MLowest
Vercel AI Gateway$0.12/M
Requesty$0.2/M
TensorX$0.2/M
Cortecs$0.201/M
Cost calculator
Estimate a workload
$0.00
Excludes taxes, non-token charges, and tiered discounts.
Context limits also differ by provider, from 262.144K to 1.04858M tokens. Compare the provider table above before choosing on price alone.
Specification
Capabilities
Recorded from the source catalog and provider listings.
Every result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.
Each axis is this model's percentile among the 7 benchmarks it has published results for, measured against every other model with a score on that same benchmark. Percentiles are used because benchmarks do not share a scale — a 60 on one is not a 60 on another. Hover any point for the raw score.
Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.
Cost against capability
Price and performance
Listed input price plotted against Agents' Last Exam, the benchmark with the widest published coverage that this model appears in.
The stepped line is the efficient frontier: at each price, the best score available for that money or less. A model sitting on it is not being beaten by anything cheaper. Price is log-scaled because listed rates span four orders of magnitude. Only models with both a listed price and a score on this benchmark can appear.
Catalog activity
Change log
Field-level changes detected between successful source imports.
Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.
What is Qwen3.8 Flash Next?
Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding. It is published by Alibaba Qwen and catalogued here from Models.dev.
How much does Qwen3.8 Flash Next cost?
Listed input pricing starts at $0.1 per million tokens from Vultr, rising to $0.201 across 5 listed providers.
What is the context length of Qwen3.8 Flash Next?
Qwen3.8 Flash Next accepts up to 262.144K tokens of context and returns up to 131.072K output tokens.
Does Qwen3.8 Flash Next support tool calling and structured output?
Provider catalogs list support for tool calling, structured output, reasoning, and image input.
Which providers serve Qwen3.8 Flash Next?
6 providers list this model: AMD, Vultr, Vercel AI Gateway, Requesty, TensorX, Cortecs.
Are the weights for Qwen3.8 Flash Next open?
Yes. The weights are published and downloadable from Hugging Face under the qwen-community-1.0 license.
When was Qwen3.8 Flash Next released?
The catalog records a release date of 2026-08-27, last verified Oct 24, 2026.
Vultr
Requesty
TensorX
Cortecs