Qwen 3 8B · Ollama

srchmnmichael/Qwen3.8-Uncensored

Qwen3.8-27B tensor-level abliterated, vision tower and MTP head untouched. 0% over-refusal on XSTest, 0-6% refusal across the A/B suite, no measurable capability loss. Full mmproj vision, tool calling and thinking, 262K context.

vision tools thinking

1,458 Pulls 4 Tags Updated 3 days ago

orcarouter/Qwen3.8-Flash-Next-Uncensored

Qwen3.8-Flash-Next tensor-level abliterated— a latest Qwen4-architecture MoE (~177B total / ~6B active) with vision, reasoning, tool-calling, and 262K context. MLX 4/6/8-bit for Apple Silicon. Research use only.

vision tools thinking

1,374 Pulls 4 Tags Updated 2 days ago

aiconjured/Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4

Text + Vision Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF-Q8-NVFP4

vision

526 Pulls 1 Tag Updated 6 days ago

ttempvnn/0bserverx-Qwen3.8-27B-Heretic-Abliterated-Uncensored-multilingual-Q4-K-M

216 Pulls 1 Tag Updated 5 days ago

thomboe/Qwen3.8-27B-Uncensored

vision

35 Pulls 1 Tag Updated yesterday

ttempvnn/JonathanColetti-Qwen3.8-27B-Uncensored-Q4-K-M

33 Pulls 1 Tag Updated yesterday

MobiusDevelopment/Qwen3.8-27B-MTP-Q4_K_M_GGUF-Uncensored

Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks.

vision

5 Pulls 1 Tag Updated an hour ago

orcarouter/Qwen3.8-27B-Uncensored

Qwen3.8-27B tensor-level abliterated, vision tower and MTP head untouched. 0% over-refusal on XSTest, 0-6% refusal across the A/B suite, no measurable capability loss. Full mmproj vision, tool calling and thinking, 262K context.

vision tools thinking

123.6K Pulls 20 Tags Updated 2 days ago

smtek/Qwen3.8-27B

Quants from Q2 up to Q5 from Unsloth K_M and UD_K_XL and builds for 12, 16 and 24 GB

vision

12.6K Pulls 22 Tags Updated 1 week ago

tinyrick/Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF

HauhauCS/Qwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF

vision

2,375 Pulls 3 Tags Updated 1 week ago

jikepjikep_16HEX/qwen3.8-27b-nightshift-heretic-uncensored-q4

😈 Uncensored Qwen3.8-27B Vision (27.3B • Q4_K_M) for Ollama. 👁️ Multimodal Vision AI with CLIP-ViT, Thinking, MTP, Tool Calling, Rust 1.98.0 Image Analysis, GGUF & local AI development. Medic🚀 #16HEX Matrix / Nightshift Heretic.

vision

2,060 Pulls 1 Tag Updated 1 week ago

ttempvnn/0bserverx-Qwen3.8-27B-Heretic-Abliterated-Uncensored-Q4-K-M

vision

1,247 Pulls 1 Tag Updated 5 days ago

Julioa/Qwen3.8-27B-UD-Q3_K_XL

658 Pulls 1 Tag Updated 1 week ago

oamazonasgabriel/qwen3.8-27b

Qwen3.8-27B in Q4_K_M quantization (32GB+ VRAM required). Dense 27.8B parameters with hybrid attention for long context (256K tokens). Apache 2.0 license. Ideal for high-VRAM setups (RTX 5090/4090 dual, M4 Ultra, etc.).

744 Pulls 3 Tags Updated 2 days ago

zhamm/Qwen3.8

High-quality Q8_0 quantization of Qwen3.8-27B packaged for Ollama with full vision support and MTP for speed.

vision

569 Pulls 1 Tag Updated 1 week ago

ttempvnn/Blackfrost-AI-Qwen3.8-27B-ABLITERATED

vision

352 Pulls 1 Tag Updated 2 weeks ago

jetelain/Qwen3.8-27B

Qwen3.8-27B UD-Q4_K_XL MTP from Unsloth

vision tools thinking

244 Pulls 1 Tag Updated 1 week ago

aiconjured/Qwen3.8-27B-NVFP4-MTP-COMPACT-LOW

Text + Vision All credit: https://huggingface.co/esatapedico/Qwen3.8-27B-NVFP4-MTP-GGUF

vision

164 Pulls 1 Tag Updated 1 week ago

Whittle/Qwen3.8-Whittle-MoE-27B-A17.8B

The architecture is a partition, not a rebuild. All 64 layers keep the parent's attention side untouched: the 3 to 1 hybrid of gated deltanet layers and full attention, 16 attention layers in all, hidden size 5120. The surgery is in the feed forward. Each

145 Pulls 4 Tags Updated 6 days ago

lucifers/qwen3.8

Qwen3.8:27B-MTP.Q6_K

165 Pulls 1 Tag Updated 2 weeks ago