Qwen3.5-0.8B-uncensored-MNN
darkmaniac7基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
언어부터 이미지, 오픈 웨이트부터 API까지. 기능과 실제 비용을 비교하세요.
다음 아이디어에 맞는 AI를 찾아보세요.
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
GGUF builds of darrellbest/Qwen-Image-2.1-PE-I2I-Heretic, the refusal-ablated image-editing prompt rewriter: it reads the image you are editing and turns a short instruction into a detailed edit prompt. For **llama.cpp** and **Ollama**, including machines with no CUDA GPU and ARM machines such as the DGX Spark. > **Not affiliated with or endorsed by Alibaba / Qwen.** Derived from > `Qwen/Qwen-Image-2.1-PE-I2I` under the **Qwen Research > License** (copy included). **Non-commercial use only.** **`system_prompt.txt` is included and required.** It defines the output contract, and the model is useless without it.
번역 대기 중 · 원문 표시
GGUF builds of the text encoder shipped in Qwen/Qwen-Image-2.1 (`text_encoder/`). That encoder is stock Qwen/Qwen3-VL-8B-Instruct: all 750 tensors match in name, shape and dtype, and every sampled tensor is byte-identical. So these files also work as a general Qwen3-VL-8B-Instruct GGUF. **Q6_K and Q4_K_M are not stock recipes.** Plain Q6_K visibly garbled rendered text, so both use: - an importance matrix computed on ~1500 real image prompts wrapped in the pipeline's own encoder template (`llama-imatrix --parse-special`); - the last 4 decoder layers (32-35) kept at Q8_0, because the pipeline reads the last layer's output directly; - token embeddings at Q8_0. **Recommendation:** Q8_0 if it fits. Q6_K is the smallest build that rendered text correctly in testing. Q4_K_M is usable, but at the same seed it changes composition and can misspell rendered text.
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
GGUF builds of darrellbest/Qwen3.5-4B-Heretic. Qwen/Qwen3.5-4B with its refusal behaviour removed by Heretic using full-weight Arbitrary-Rank Ablation: **6/100 refusals** (original: 99/100) at **KL divergence 0.0220**. See the main repository for how it was made and measured. Converted from the bf16 safetensors with llama.cpp's `convert_hf_to_gguf.py` and quantized with `llama-quantize` (no imatrix). The mmproj file carries the vision encoder; load it alongside any of the three for image input. Every file was loaded in `llama-server` with the mmproj and in Ollama. All three answered ordinary prompts correctly (BF16 and Q8_0 reproduce each other word for word), described a test image (a red circle and a blue square) correctly, and in Ollama answered 17 x 23 = 391 with thinking off. Thinking-mode reasoning, 4 arithmetic and word problems x 10 seeds with Qwen's recommended sampling: BF16 39/40, Q8_0 40/40, Q4_K_M 40/40 finished and correct (the original model in vLLM: 40/40).
번역 대기 중 · 원문 표시
GGUF builds of darrellbest/Qwen3.5-9B-Heretic. Qwen/Qwen3.5-9B with its refusal behaviour removed by Heretic using full-weight Arbitrary-Rank Ablation: **5/100 refusals** (original: 100/100) at **KL divergence 0.0403**. See the main repository for how it was made and measured. Converted from the bf16 safetensors with llama.cpp's `convert_hf_to_gguf.py` and quantized with `llama-quantize` (no imatrix). The mmproj file carries the vision encoder; load it alongside any of the three for image input. Every file was loaded in `llama-server` with the mmproj and in Ollama. All three answered ordinary prompts correctly, described a test image (a red circle and a blue square) correctly, and in Ollama answered 17 x 23 = 391 with thinking off. Thinking-mode reasoning, 4 arithmetic and word problems x 10 seeds with Qwen's recommended sampling: BF16 39/40, Q8_0 37/40, Q4_K_M 35/40 finished and correct (the original model in vLLM: 40/40); the misses ran out of the 4,000-token budget rather than answering wrongly.
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
GGUF builds of darrellbest/Qwen3.8-27B-Heretic: Qwen3.8-27B with its refusal behaviour removed (0/100 refusals vs 98/100 for the original, KL divergence 0.0465). The three model files are text-only: this is a vision-language model, and llama.cpp keeps the vision encoder in the separate `mmproj` file. Add it only if you want to send images. The included `Modelfile` sets `RENDERER qwen3.8` and `PARSER qwen3.5`, Ollama's built-in support for this architecture. They matter: without them the model's reasoning text ends up inside the answer, and generation was measured about half as fast (9.5 vs 18.8 tokens/s on an RTX PRO 6000 with Q4_K_M).
번역 대기 중 · 원문 표시
Chandra is an OCR model that outputs markdown, HTML, and JSON. It is highly accurate at extracting text from images and PDFs, while preserving layout information. You can try Chandra in the free playground here, or at a hosted API here. - Convert documents to markdown, html, or json with detailed layout information - Good handwriting support - Reconstructs forms accurately, including checkboxes - Good support for tables, math, and complex layouts - Extracts images and diagrams, with captions and structured data - Support for 40+ languages
번역 대기 중 · 원문 표시
lift is a structured extraction model from Datalab that pulls structured JSON out of PDFs and images. Pass any JSON schema and lift returns a JSON object matching it, using schema-constrained decoding to guarantee valid, well-typed output. Try lift in the free playground, or use the hosted API for higher accuracy, per-field verification, and citations. - Extract structured data from documents - Pass any JSON schema - Handles multi-page documents in a single pass, including values that span pages - Two inference modes: local (HuggingFace) and remote (vLLM server) - CLI for single files, inline schemas, or whole directories - Schema Studio: a Streamlit app to build, save, and test schemas against your documents
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
- Accuracy - scores 83.3% on olmOCR-bench (top under 3B params) - Speed - throughput of 5 pages/s on an RTX 5090 - Multilingual - scores 87.2% on an internal benchmark set of 91 languages (more here) - Layout analysis (table, image, header, etc.) with reading order - Table recognition (rows + columns) It works on a range of documents (see usage and benchmarks). Our managed platform runs both Surya, and variants of our highest accuracy model, Chandra.
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
**The Coder, uncensored.** This release takes ISTA-DASLab's pruned **GSQ-RCO Coder** (IQ1_M tier, 256/512 experts per layer) and swaps its refusal behavior for the **abliterated (uncensored) core** of `SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF` — as a **weights-only transplant**. No runtime control vector is needed: the abliteration is in the tensors. ("Uncen" here means exactly that abliterated core; the original build was prepared under the name *Coder-Abliterated*, which `MANIFEST.json` keeps.) > **Runtime support.** Built for and served with the **Strata** engine. Stock llama.cpp / Ollama / vLLM are not tested with this artifact. Behavior reports are welcome. Please open a **discussion in the Community tab** with: hardware + RAM, engine + version, context/settings, and what you observed (refusals or their absence, behaviour, languages).
번역 대기 중 · 원문 표시
> ⚠️ **Read this first — ignore the automatic "Use this model" panel.** It suggests a `:BF16` llama.cpp/Ollama launch that does not match this release. This GGUF runs with the **Strata** engine: follow the Quick start below (stock llama.cpp / Ollama / vLLM are not tested). **Swift-1.5 Uncensored (`Uncen`)** is an abliterated (refusal-reduced) build of Swift 1.5 — the uncensored line of the Qwen3.8-Flash-Next GSQ-RCO family. This release shrinks the MoE expert pool from **512 → 332 experts/layer** so it serves on consumer hardware (a single 16 GB GPU + 32 GB RAM with the Strata engine), while keeping as much coding, agent and tool-calling capability as possible. - **About the `IQ1` name — read first.** `IQ1` here does **not** mean 1-bit tensor quantization. It labels the **effective payload class** of the pruned release — ~60.65 GB (56.49 GiB) — the same naming convention and size class as the GSQ-RCO Coder `IQ1_M` release (58.4 GB; itself a pruned build). Every tensor in this release keeps its original quantization inherited from the source; nothing was re-quantized. Equal size class is **not** quantization equivalence: this release keeps the source's mixed `IQ3_XXS` tensor bytes, while the Coder's kept experts are stored at ~3.5 bpw — read neither name as a tensor-precision tier. - **Base:** `SC117/Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF` @ `46d8cfdd` (IQ3_XXS tier) — all weights, including every kept expert, come from this uncensored source. - **Experts:** 512 → **332 per layer** (≈ 64.8% of the pool), 48 layers, 10 routed experts/token. - **Payload:** 36.47 GB + 24.18 GB = **60.65 GB** (56.49 GiB) across 2 GGUF shards; the vision projector is included. - **Verified setup (author's experience):** Windows 11 · single **RX 6800 16 GB** + **32 GB system RAM** · Strata engine with `--resident-experts` and expert caching · KV int8 · MTP speculative decoding. Production-served at 144K–192K context; 256K context was verified at runtime (a ~251.5K-token prompt was recalled correctly in testing). Other hardware will vary. - **Multilingual:** inherits Qwen3.8-Flash-Next's language coverage. Smoke-tested in **eight languages** across Latin, CJK and Cyrillic scripts — arithmetic + explanation passed in all eight; free-form writing is shown for the languages that met the bar (see `i18n-smoke/`). - **Vision:** image-text input works with the included proje
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
A **W4A16 (AWQ)** quantization of `Blackfrost-AI/Qwen3.8-Flash-Next-DERISKED-BF16`, an abliterated ("derisked") variant of Qwen's Qwen3.8-Flash-Next. Multimodal (image-text-to-text), MTP speculative-decode head and PLE n-gram table preserved. Built to run the 80B-class Flash-Next MoE on 2× AMD MI210 (gfx90a) at long context. - **Format:** `compressed-tensors`, **W4A16** — 4-bit weights, group size **128**, symmetric, no zero-point (loads on vLLM's `TritonW4A16` + `fused_moe_kernel_gptq_awq` fast paths). - **Method:** **AWQ** (activation-aware). Smoothing is restricted to the mathematically-exact `up_proj → down_proj` pairs. The standard `norm → q/k/v` and `norm → gate/up` smoothings are deliberately **excluded**: this architecture uses hyper-connections with a data-dependent (sigmoid) gate, so folding a smoothing scale into the norm would change the gate non-linearly and corrupt the weights. - **Quantized:** routed experts (gate/up/down) and full-attention `q/k/v/o`. - **Kept in bf16** (matching the reference recipe): the GDN linear-attention layers (recurrent state — error compounds along the sequence), the shared expert, the QSA indexer, the router gates, hyper-connections, norms, `embed_tokens`, `lm_head`, the **MTP** speculative-decode head, and the **PLE** n-gram embedding table.
번역 대기 중 · 원문 표시
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
基础档案已收录,详细介绍后续补充。可先查看厂商、分类与来源。
기본 정보 등록 · 상세 정보 보완 예정
제공사, 분류, 입출력 조건을 조정하세요. 새 모델이 계속 등록됩니다.
월간 토큰 사용량으로 예상 API 비용을 비교하세요.
수집된 백만 토큰당 USD 요금입니다. 캐시 할인, 구간 요금, 추가 추론 출력, 세금은 제외됩니다. 최신 요금은 출처에서 확인하세요.
입력, 출력, 캐시 요금을 구분하고 동일 작업량으로 비교하세요. 이미지, 음성, 영상은 개수, 시간, 해상도로 과금될 수 있습니다.
반드시 그렇지는 않습니다. 버전별 라이선스, 상업 이용 제한, 정책을 확인하세요. 자체 호스팅에는 연산 및 유지 비용도 듭니다.
순위는 작업, 버전, 평가 방식에 따라 다릅니다. 실제 업무로 정확도, 지연, 안정성, 비용을 비교하세요.