iFLYTEK releases Spark X2.6, a 293B-A30B MoE model trained on domestic compute
At its 1024 developer festival, iFLYTEK released Spark X2.6, a 293B-A30B mixture-of-experts model trained on domestic compute. The company also open-sourced the on-device Spark X2.6-4B and debuted the SparkAudio 1.0 speech model.
iFLYTEK has released Spark X2.6, a 293B-A30B mixture-of-experts (MoE) model trained on domestic compute. The company announced it at its 1024 developer festival, as reported by Pandaily.
What was announced
- Spark X2.6: a 293B-A30B MoE model, trained on domestic compute.
- Spark X2.6-4B: an on-device model, open-sourced by iFLYTEK.
- SparkAudio 1.0: a speech model, debuted at the same event.
The "293B-A30B" label follows the common MoE naming convention, which pairs a total parameter count with the number of parameters active per token.
What this means
iFLYTEK is releasing three things at once: a large MoE model trained on domestic compute, an open-source 4B model meant to run on devices, and a new speech model. The open-sourced Spark X2.6-4B is the part developers can build on directly.
Related Articles
Xiaomi open-sources Xiaomi-Lens-0, a 4B multimodal model that outputs pixel-level mask tokens mid-answer
Xiaomi's SeerRay-Lab has open-sourced Xiaomi-Lens-0, a 4B multimodal model built on Qwen3.5-4B. The model emits mask tokens for pixel-level image regions in the middle of an answer. The lab also released a benchmark called LensBench.
Recursive releases Flint, a 40B-parameter open coding model that runs on laptops under Apache 2.0
Recursive released Flint, a free, open-weight mixture-of-experts coding model with 40 billion total and 2.5 billion active parameters, under Apache 2.0. The company claims its Eureka Machine automated research system contributed to nearly every stage of development, and that Flint runs at about 131 tokens per second on an M4 Pro MacBook Pro.
NVIDIA releases Kumo Forecast, an Apache 2.0 time-series model built on a ~340M-parameter MOMENT-1-large encoder
NVIDIA has released Kumo Forecast (formerly NV-Tesseract Forecasting), a multivariate time-series forecasting model under Apache 2.0. It pairs a frozen MOMENT-1-large encoder of roughly 340 million parameters with a trainable forecasting head, optional nearest-neighbor retrieval over historical data (DARR), and a built-in interpretability framework.
Hugging Face releases Carbon-X, a 14B-parameter open model that answers natural-language questions about DNA
Hugging Face released Carbon-X, a 14-billion-parameter open model that takes DNA sequences and English questions and returns conversational answers. It was trained on 300 billion tokens and is designed to run locally. Its developers say it has not yet been validated with wet lab experiments.
Comments
Loading...