Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.
C++
Merge f-020-docs2 into f-020-int (0.2.0 release wiring and docs)
Niko1221e46be1fe0af2
diff --cc docs/INTEL.mdindex 915d2379c,8b89f9e4f..eb18988d1--- a/docs/INTEL.md+++ b/docs/INTEL.md@@@ -770,16 -770,27 +770,33 @@@ teacher-forced KL against the default ois 0.012), first-token KL 0.0017 (floor 0.0005), top-1 agreement 99.7%. HumanEval (every 2nd, 82) 73, 69 and 74 for the mode and two default runs, reasoning 21 / 23 / 22 of30, a 35-question Chinese/Japanese/Korean set 35/35 in all, needles at 2K-6K 9/9, an 8K-context tool-call agent 6/6.+ ### A750: a prompt that never finished is fixed (engine 0.1.45.3, commit `fca3c5958`)++ On an Arc A750 some prompts never finished: with default flags a 2,744-token documentation excerpt hung in 6 of 6 runs (the+ thread spun in `sched_yield` inside `libze_intel_gpu` with every engine idle), and the same hang was seen on a 5.3K-token needle prompt. The+ cause: SYCL has no `cudaHostRegister`, so the expert arena is ordinary pageable memory even where the engine calls it+ pinned. Copying a 2.3 MB expert straight out of it went through the Level Zero driver's staging path on the calling thread,+ and on some prompts that call never returned. Now only experts in USM host memory (the `--stream-experts` pinned mirror)+ are copied directly (the answer is remembered per 2 MiB); every other expert goes through the stager's pinned buffers, like+ an unpinned one. **`STRATA_PF_DMA_PAGEABLE=1` restores the old direct copy.**++ Checked: the 2,744-token repro finished 3 of 3 times after the fix (0 of 6 before); the first-token logits and the tokens are+ identical old against new on the A750 (1.4K, 2.5K and 4K prompts) and on the Arc Pro B70 (stock config, and without+ `--stream-experts`); speed is unchanged where it did not hang (A750 4K prompt 79.4 against 79.3 tok/s, B70 stock 1,246+ against 1,240), and on the B70 without `--stream-experts` the prompt is faster (680-740 against 77-456 tok/s). The+ regression test is `sycl/tools/test_prompt_deadlock.sh` (the 2,744-token prompt; it times out on the old build and passes on+ the fix).+## Not done-- Images, on both Intel engines. Strata's vision path encodes with `strata-vision` into embeddings the CUDA- engine reads; neither the SYCL port nor the llama.cpp config wires it yet.+- Images on the llama.cpp config: not wired. On the SYCL port pictures are wired but opt-in (`setup --backend sycl --vision yes`):+ setup compiles the portable CPU reader `strata-vision` (the program the AMD packages carry beside their engine) into `engine-sycl/`,+ the server runs it on the host, and the engine in the container reads its embeddings from the host's `/dev/shm`, which+ `strata-sycl.sh` shares (`STRATA_SYCL_SHM=0` keeps a private one). A SYCL GPU reader does not exist yet, so every Arc reads+ pictures on the CPU (a few seconds each). They are not on by default because the first picture request on the Arc Pro B70+ (engine 0.1.45, IQ3_S, 2026-10-19) hung the GPU: the xe driver logged an engine reset and `Fault response: Unsuccessful -EINVAL`+ and the request ended 503. Text was not affected by this wiring (the reader and the shared `/dev/shm` are only used for pictures);+ the engine's vision path on SYCL needs a look first (`setup_intel.PICTURES_ON_BY_DEFAULT` flips the default).- Speculative decoding on the llama.cpp path (the GGUF carries no draft layer llama.cpp can use). The SYCLport has it (MTP draft layer, `--mtp`).