Spark-X2.6-4B (4.11 B dense, hybrid attention: 9 full layers + 27 sliding-window layers of 512 tokens, 1M-token context, thinking mode, tool calling) packaged as single CMF files. They run with cortiq, a Rust inference engine with no Python and no ML framework: NVIDIA, AMD and Intel GPUs through Vulkan/DX12, Apple silicon through Metal, and the CPU. X2.6 keeps the Spark-X2.5-4B architecture and improves post-training (coding, agentic, long-context, multilingual, reasoning). Use cortiq 0.8.43 or later; prebuilt binaries for Linux, macOS and Windows are on the releases page. The engine picks the GPU and its settings itself. —no-think asks for a direct answer; without it the model reasons in a block first, so raise —max-tokens (default 256). Both files carry 362 tensors: 253 quantized projections + 109 f16 (embeddings, norms, rotary buffers), quantized from the bf16 checkpoint. Perplexity: first 4096 tokens of wikitext-2 test with the model’s BOS token in front, cortiq ppl —tokens 4096, exact attention, measured on the RTX 2000 Ada. Cross-check with the identical command on Spark-X2.5-4B-q82f.cmf` gives 10.94 (that card lists 10.68; the text file here is rebuilt from the parquet,…
Модальности:
Генерация текста
Области применения:
Мультиязычность
Задача: Генерация текста
Автор: infosave
Теги: cortiq, cmf, quantized, sliding-window-attention, tool-calling, multilingual, en, zh
Лайков: 3 | Загрузок: 586
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
