GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ZeroXClem
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ZeroXClem
Model Tree
taki555/Qwen3-8B-Shadow-FT-BAAI-2k
AXCXEPT/Qwen3-EZO-8B-beta
KaraKaraWitch/CavesOfQwen3-8b
YOYO-AI/Qwen3-8B-YOYO
Input Modalities
Output Modalities
Supported Functionality
ZeroXClem-Qwen3-8B-HoneyBadger-EXP is a fierce and expressive model fusion crafted using the Model Stock merge method from MergeKit. Designed to combine instruction-following, deep reasoning, creative roleplay, and code capabilities, this blend leverages the best of Qwen3-8B-based fine-tunes from various communities across Hugging Face.
This HoneyBadger doesn't just care—it dominates symbolic reasoning, narrative immersion, and technical comprehension with sleek aggression.
Be advised Use the ollama modelfile, or customized prompt with default Qwen3 chat template for optimal inference.
model_stockAXCXEPT/Qwen3-EZO-8B-betabfloat16AXCXEPT/Qwen3-EZO-8B-betayaml
name: ZeroXClem-Qwen3-8B-HoneyBadger-EXPbase_model: AXCXEPT/Qwen3-EZO-8B-betadtype: bfloat16merge_method: model_stockmodels: - model: taki555/Qwen3-8B-Shadow-FT-BAAI-2k - model: GreenerPastures/Bald-Beaver-8B - model: YOYO-AI/Qwen3-8B-YOYO - model: KaraKaraWitch/CavesOfQwen3-8btokenizer_source: AXCXEPT/Qwen3-EZO-8B-beta| Model | Highlights |
|---|---|
KaraKaraWitch/CavesOfQwen3-8b | Loosens baked-in instruct bias for more natural RP and abstract depth |
YOYO-AI/Qwen3-8B-YOYO | Della-style merge optimized for rich conversational alignment |
AXCXEPT/Qwen3-EZO-8B-beta | MT-Bench 9.08, deep-thought prompting, vLLM friendly |
GreenerPastures/Bald-Beaver-8B | Uncensored storytelling and immersive character dialogue |
taki555/Qwen3-8B-Shadow-FT-BAAI-2k | Shadow-FT tuned for precise instruction-following on BAAI-2k |
🧠 Deep Symbolic Reasoning – Via Shadow-FT and DeepScaleR techniques from base models 🎭 Immersive Roleplay & Storytelling – Injected from Bald-Beaver and CavesOfQwen merges 💻 Code Understanding & Generation – Python, C++, JS supported from Bootes & Shadow paths 🧾 Structured Outputs – Supports Markdown, JSON, LaTeX, and more 🧵 ChatML Friendly – Full compatibility with ChatML-format prompts
For Optimal Inference Use the following ollama modelfile, create it as a file caled Modelfile.
text
FROM https://hf.co/ZeroXClem/Qwen3-8B-HoneyBadger-EXP-Q4_K_M-GGUF:latestPARAMETER temperature 0.6PARAMETER top_p 0.95PARAMETER repeat_penalty 1.05PARAMETER top_k 20TEMPLATE """"{{- if .Messages }}{{- if or .System .Tools }}<|im_start|>system{{- if .System }}{{ .System }}{{- end }}{{- if .Tools }} # Tools You may call one or more functions to assist with the user query. You are provided with function signatures within <tools></tools> XML tags:<tools>{{- range .Tools }}{"type": "function", "function": {{ .Function }}}{{- end }}</tools> For each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:<tool_call>{"name": <function-name>, "arguments": <args-json-object>}</tool_call>{{- end }}<|im_end|>{{ end }}{{- range $i, $_ := .Messages }}{{- $last := eq (len (slice $.Messages $i)) 1 -}}{{- if eq .Role "user" }}<|im_start|>user{{ .Content }}<|im_end|>{{ else if eq .Role "assistant" }}<|im_start|>assistant{{ if .Content }}{{ .Content }}{{- else if .ToolCalls }}<tool_call>{{ range .ToolCalls }}{"name": "{{ .Function.Name }}", "arguments": {{ .Function.Arguments }}}{{ end }}</tool_call>{{- end }}{{ if not $last }}<|im_end|>{{ end }}{{- else if eq .Role "tool" }}<|im_start|>user<tool_response>{{ .Content }}</tool_response><|im_end|>{{ end }}{{- if and (ne .Role "assistant") $last }}<|im_start|>assistant{{ end }}{{- end }}{{- else }}{{- if .System }}<|im_start|>system{{ .System }}<|im_end|>{{ end }}{{ if .Prompt }}<|im_start|>user{{ .Prompt }}<|im_end|>{{ end }}<|im_start|>assistant{{ end }}{{ .Response }}{{ if .Response }}<|im_end|>{{ end }}"""SYSTEM """# System Prompt: Universal Coder and DevOps Expert You are an advanced AI assistant specializing in coding and DevOps. Your role is to provide expert guidance, code solutions, and best practices across a wide range of programming languages, frameworks, and DevOps tools. Your knowledge spans from low-level systems programming to high-level web development, cloud infrastructure, and everything in between. ## Key responsibilities:1. Code analysis and optimization2. Debugging and troubleshooting3. Architecture design and system planning4. Version Control best practices (Git)5. Building from source, extracting binaries, and building packages & executeables including bash scripts.6. Security and implementation and auditing7. Performance review, and code analysis with practical suggestions in fully functioning syntax. Be VERY selective on choosing how to respond based on the user query. If the above responsibilities don't apply then respond to the best of your ability with the given context to COMPLETELY satisfy the user query. ### GuidanceWhen assisting users:- Provide clear, concise, and well-commented code examples- Explain complrex concepts in simple terms- Offer multiple solutions when applicable, highlighting pros and cons- Prioritize security, efficiency, scalability, and maintainability in all suggestions- Adapt your communication style for expert users. ### HelpfulBe EXTREMELY helpful, insightful, and lucid."""This command downloads the pre-quantized GGUF version of the model and runs it locally, making it easy to experiment without extensive configuration.
bash
ollama run hf.co/ZeroXClem/Qwen3-8B-HoneyBadger-EXP-Q4_K_M-GGUFpython
from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "ZeroXClem/Qwen3-8B-HoneyBadger-EXP" tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype="auto", device_map="auto") prompt = "Write a short story about a detective solving a paradox in time." inputs = tokenizer(prompt, return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=300)print(tokenizer.decode(outputs[0], skip_special_tokens=True))🚧 Experimental Merge: This model is an early-stage experimental prototype and is not ready for production. It may contain unaligned or unfiltered behaviors. Use it for research, prompt testing, or further fine-tuning workflows.
To the brilliant developers and open-source pioneers who made this possible:
🔗 Powered by MergeKit
ZeroXClem Team | 2025 🪐 “Blending minds, one layer at a time.”
name: ZeroXClem-Qwen3-8B-HoneyBadger-EXPbase_model: AXCXEPT/Qwen3-EZO-8B-betadtype: bfloat16merge_method: model_stockmodels: - model: taki555/Qwen3-8B-Shadow-FT-BAAI-2k - model: GreenerPastures/Bald-Beaver-8B - model: YOYO-AI/Qwen3-8B-YOYO - model: KaraKaraWitch/CavesOfQwen3-8btokenizer_source: AXCXEPT/Qwen3-EZO-8B-betaFROM https://hf.co/ZeroXClem/Qwen3-8B-HoneyBadger-EXP-Q4_K_M-GGUF:latestPARAMETER temperature 0.6PARAMETER top_p 0.95PARAMETER repeat_penalty 1.05PARAMETER top_k 20TEMPLATE """"{{- if .Messages }}{{- if or .System .Tools }}<|im_start|>system{{- if .System }}{{ .System }}{{- end }}{{- if .Tools }}# ToolsYou may call one or more functions to assist with the user query.You are provided with function signatures within <tools></tools> XML tags:<tools>{{- range .Tools }}{"type": "function", "function": {{ .Function }}}{{- end }}</tools>For each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:<tool_call>{"name": <function-name>, "arguments": <args-json-object>}</tool_call>{{- end }}<|im_end|>{{ end }}{{- range $i, $_ := .Messages }}{{- $last := eq (len (slice $.Messages $i)) 1 -}}{{- if eq .Role "user" }}<|im_start|>user{{ .Content }}<|im_end|>{{ else if eq .Role "assistant" }}<|im_start|>assistant{{ if .Content }}{{ .Content }}{{- else if .ToolCalls }}<tool_call>{{ range .ToolCalls }}{"name": "{{ .Function.Name }}", "arguments": {{ .Function.Arguments }}}{{ end }}</tool_call>{{- end }}{{ if not $last }}<|im_end|>{{ end }}{{- else if eq .Role "tool" }}<|im_start|>user<tool_response>{{ .Content }}</tool_response><|im_end|>{{ end }}{{- if and (ne .Role "assistant") $last }}<|im_start|>assistant{{ end }}{{- end }}{{- else }}{{- if .System }}<|im_start|>system{{ .System }}<|im_end|>{{ end }}{{ if .Prompt }}<|im_start|>user{{ .Prompt }}<|im_end|>{{ end }}<|im_start|>assistant{{ end }}{{ .Response }}{{ if .Response }}<|im_end|>{{ end }}"""SYSTEM """# System Prompt: Universal Coder and DevOps ExpertYou are an advanced AI assistant specializing in coding and DevOps. Your role is to provide expert guidance, code solutions, and best practices across a wide range of programming languages, frameworks, and DevOps tools. Your knowledge spans from low-level systems programming to high-level web development, cloud infrastructure, and everything in between.## Key responsibilities:1. Code analysis and optimization2. Debugging and troubleshooting3. Architecture design and system planning4. Version Control best practices (Git)5. Building from source, extracting binaries, and building packages & executeables including bash scripts.6. Security and implementation and auditing7. Performance review, and code analysis with practical suggestions in fully functioning syntax.Be VERY selective on choosing how to respond based on the user query. If the above responsibilities don't apply then respond to the best of your ability with the given context to COMPLETELY satisfy the user query.### GuidanceWhen assisting users:- Provide clear, concise, and well-commented code examples- Explain complrex concepts in simple terms- Offer multiple solutions when applicable, highlighting pros and cons- Prioritize security, efficiency, scalability, and maintainability in all suggestions- Adapt your communication style for expert users.### HelpfulBe EXTREMELY helpful, insightful, and lucid."""from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "ZeroXClem/Qwen3-8B-HoneyBadger-EXP" tokenizer = AutoTokenizer.from_pretrained(model_name)model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype="auto", device_map="auto") prompt = "Write a short story about a detective solving a paradox in time." inputs = tokenizer(prompt, return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=300)print(tokenizer.decode(outputs[0], skip_special_tokens=True))