Qwen3.6 35B A3B Heretic NVFP4

--- license: apache-2.0 language: - en library_name: transformers pipeline_tag: image-text-to-text base_model: tvall43/Qwen3.6-35B-A3B-heretic base_model_relation: quantized tags: - 1m-context - 35b - 3b-active - 4-bit - aarch64 - abliterated - aeon - aeon-7 - agentic - arm64 - blackwell - chat - chunked-prefill - coding - compressed-tensors - conversational - decensored - dflash - dgx-spark - english - flashinfer-cutlass - fp4 - fp8-kv-cache - function-calling - gated-deltanet - gb10 - gdn - gpu - grace-blackwell - heretic - hybrid - hybrid-attention - image-text-to-text - instruct - linear-attention - llmcompressor - long-context - mixture-of-experts - moe - multimodal - native-fp4 - nvfp4 - openai-api - openai-compatible - prefix-caching - production-ready - quantized - qwen - qwen3 - qwen3.5 - qwen3.6 - qwen3_5_moe - reasoning - refusal-removed - safetensors - sm_121a - sparse - sparse-moe - speculative-decoding - thinking - tool-calling - uncensored - unfiltered - vision - vision-language - vllm - weight-quantization quantization: nvfp4 --- # Qwen3.6-35B-A3B-heretic — NVFP4 (v2 multimodal-preserved)  > ✅ **Validated on the unified [AEON vLLM Ultimate image](https://github.com/AEON-7/vllm-ultimate-dgx-spark) `ghcr.io/aeon-7/aeon-vllm-ultimate:latest`** (= `:2026-06-18-v0.23.0-dflashfix`; rollback `:2026-06-11-pr41703`) — vLLM 0.23.0 built from source for GB10 / sm_121a + the AEON DFlash stack. Loads + serves cleanly (compressed-tensors MoE) — **including working multimodal image understanding** (a vision tensor-prefix bug was fixed 2026-06-18 — see the **Vision fix** section below) — and with DFlash `num_speculative_tokens: 11` now scales to **c=64** with **up to ~740 tok/s aggregate** and **no concurrency crash**. This single image supersedes the older `vllm-spark-omni-q36:v1.2` container. **Recommended container base.** > **🚀 PRODUCTION DEPLOYMENT GUIDE: [github.com/AEON-7/Qwen3.6-35B-A3B-heretic-NVFP4-DFlash](https://github.com/AEON-7/Qwen3.6-35B-A3B-heretic-NVFP4-DFlash)** > > The GitHub repo is the **definitive turn-key setup** for DGX Spark — end-to-end > deployment guide, validated OpenClaw config, and a concurrency-sweep benchmark harness. > > - Image: `ghcr.io/aeon-7/aeon-vllm-ultimate:latest` (vLLM 0.23.0 source-built for sm_121a + the AEON speculative-decoding stack — see...