Model overview
This is a GGUF-quantized version of Qwen3.8-27B produced with **ByteShape's ShapeLearn**, which learns the optimal datatype per tensor to maintain high quality even at very low bitlengths. This release replaces our earlier ShapeLearn-Lite quantizations of Qwen3.8-27B. Every model here has been through our **full multi-benchmark evaluation suite**, covering math, coding, general knowledge, instruction following and agentic tool use in both thinking and instruct modes, scored against the BF16 baseline. To learn more about ShapeLearn and to see detailed benchmarks across GPUs, please visit our blog.
The summary below is compiled from sources by an AI model, not our own testing. Verify selection decisions against original sources and your workload.
Key features
- GGUF-quantized version of Qwen3.8-27B optimized via ByteShape's ShapeLearn, which assigns optimal datatypes per tensor for quality preservation at low bitlengths
- Replaces earlier ShapeLearn-Lite quantizations of the same base model
- Evaluated across a full multi-benchmark suite covering math, coding, general knowledge, instruction following, and agentic tool use in both thinking and instruct modes, scored against BF16 baseline
Limitations & considerations
Sources are insufficient to establish specific limitations. Verify task performance, latency, data policies, licensing and version lifecycle.
Source prices & cost analysis
Prices vary by channel, region and configuration. Channel quotes are listed independently and are not treated as a single official provider price.
| Source / channel | Input | Output | Billing unit | Collected at |
|---|---|---|---|---|
| No verified prices yet. Check the source channel. | ||||
Estimate monthly base token costs
Estimates exclude cache discounts, tiered pricing, multimodal surcharges, web search, taxes and additional reasoning output.
Public benchmarks & run configuration
暂无已采集的可核验公开评分。模型未上榜不代表能力较低。
View full indexed rankings →Frequently asked questions
What quantization method is used for this model?
It uses ByteShape's ShapeLearn quantization, which learns the optimal datatype per tensor to maintain high quality even at very low bitlengths.
How does this quantization compare to previous versions?
This release replaces the earlier ShapeLearn-Lite quantizations of Qwen3.8-27B.
What benchmarks were used to evaluate this model?
It was evaluated on a full multi-benchmark suite covering math, coding, general knowledge, instruction following, and agentic tool use—in both thinking and instruct modes—using the BF16 baseline as reference.
Do open weights permit free commercial use?
Not necessarily. Check the version license, commercial conditions and usage policies. Self-hosting incurs compute and maintenance costs.
Are these real-time billing prices?
These are timestamped source snapshots, not live billing prices. Check the channel for current versions, regions, discounts and billing methods.
Sources & traceability
许可证:apache-2.0。模型介绍与第三方评测均以原来源的具体版本和方法为准。