Huggingface
AI inference provider available on Antbase
OpenAI-compatible
About Huggingface
Huggingface is an AI model provider accessible through Antbase. Use intelligent routing to automatically select the best model from this provider based on your task requirements.
Live Performance Data
Huggingface Models on Antbase
| Model | Score | Requests | Success Rate | Avg Latency | p95 Latency | Capabilities |
|---|---|---|---|---|---|---|
| Llama 4 Maverick 17B 128E Instruct | 78.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen2.5 72B Instruct | 76.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek R1 Distill Llama 70B | 74.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| Qwen2.5 VL 72B Instruct | 74.0 | 0 | 0.0% | 400ms | 1000ms | 👁 |
| Llama 4 Scout 17B 16E Instruct | 72.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| QwQ 32B | 72.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| Qwen2.5 Coder 32B Instruct | 70.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Llama 4 Scout | 63.0 | 192 | 10000.0% | 4904ms | 10434ms | |
| DeepSeek V3.1 | 62.0 | 254 | 10000.0% | 5671ms | 17904ms | |
| Mistral Small 3.1 | 62.0 | 1 | 10000.0% | 1657ms | 1657ms | |
| GPT-4o Mini (AskCodi) | 62.0 | 4 | 10000.0% | 1883ms | 2583ms | |
| Kimi K3 | 59.0 | 1 | 10000.0% | 2924ms | 2924ms | |
| GPT-OSS 20B | 59.0 | 1 | 10000.0% | 3086ms | 3086ms | |
| OpenAI GPT-4.1 | 58.0 | 1 | 10000.0% | 3406ms | 3406ms | |
| Gemini 2.5 Flash | 57.0 | 27 | 10000.0% | 2755ms | 8254ms | |
| Command A (2025) | 51.0 | 3 | 10000.0% | 5818ms | 13831ms | |
| Qwen3 Coder Next | 46.0 | 7 | 10000.0% | 9462ms | 12749ms | 🔧 |
| Tiny Aya Global | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Kimi K2 Instruct 0905 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.7 Flash | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| MiniMax M2.1 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 Coder 30B A3B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| GLM 4.5 Air | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen2.5 7B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 4B Thinking 2507 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| Kimi K2 Thinking | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| GLM 4.7 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Olmo 3 7B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen2.5 VL 7B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| Qwen3 Next 80B A3B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Rnj 1 Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen2.5 Coder 7B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 VL 30B A3B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek V3 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 Coder Next FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek R1 0528 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| Gemma 3n E4B It | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 4B Instruct 2507 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| C4ai Command A 03 2025 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Hermes 2 Pro Llama 3 8B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Tiny Aya Fire | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| DeepSeek R1 Distill Qwen 32B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| DeepSeek R1 Distill Qwen 7B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| DeepSeek V3.2 Exp | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Molmo2 8B | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.6V Flash | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 235B A22B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 14B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Olmo 3.1 32B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| DeepSeek V3.1 Terminus | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Trinity Mini | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 30B A3B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| L3 8B Stheno V3.2 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command A Reasoning 08 2025 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.5V | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| MiniMax M2 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Meta Llama 3 70B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| C4ai Command R7b 12 2024 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek Prover V2 671B | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 VL 235B A22B Thinking | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| GLM 4 32B 0414 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| C4ai Command R 08 2024 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Apriel 1.6 15b Thinker | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Gpt Oss Safeguard 20b | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Aya Vision 32b | 45.0 | 0 | 0.0% | 400ms | 1000ms | 👁 |
| GLM 4.6 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 Coder 480B A35B Instruct FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Olmo 3.1 32B Think | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command A Translate 08 2025 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.5V FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 235B A22B Thinking 2507 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| Llama Guard 4 12B | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Apertus 8B Instruct 2509 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| ERNIE 4.5 21B A3B PT | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 235B A22B Instruct 2507 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek R1 Distill Llama 8B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| Qwen3 VL 30B A3B Thinking | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| L3 70B Euryale V2.1 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Tiny Aya Earth | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| C4ai Command R7b Arabic 02 2025 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| ERNIE 4.5 VL 424B A47B Base PT | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| L3 8B Lunaris V1 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| ERNIE 4.5 VL 28B A3B PT | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Cogito 671b V2.1 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| MiniMax M1 80k | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 Next 80B A3B Thinking | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| Arch Router 1.5B | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Cogito 671b V2.1 FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| DictaLM 3.0 24B Thinking | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠 |
| Kimi K2 Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 VL 235B A22B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Tiny Aya Water | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| DeepSeek V3.1 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen2.5 Coder 3B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| GLM 4.6 FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.6V FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.7 FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command A Vision 07 2025 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 👁 |
| GLM 4.5 Air FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Aya Expanse 32b | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Llama 3.3 Swallow 70B Instruct V0.4 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| GLM 4.5 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Llama 3.1 70B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| EuroLLM 22B Instruct 2512 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Apertus 70B Instruct 2509 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen SEA LION V4 32B IT | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Llama 4 Maverick 17B 128E Instruct FP8 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 32B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| DeepSeek R1 Distill Qwen 1.5B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| MiniMax M2.5 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| GLM 4.6V | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| AutoGLM Phone 9B Multilingual | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3.5 9B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| WizardLM 2 8x22B | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| ERNIE 4.5 300B A47B Base PT | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3 VL 8B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Gemma SEA LION V4 27B IT | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3.5 35B A3B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3.5 27B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| GLM 5 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Qwen3.5 397B A17B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Kimi K2.5 | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Gpt Oss 120b | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| DeepSeek V3.2 | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Llama 3.1 8B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3.5 122B A10B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Meta Llama 3 8B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| MiMo V2 Flash | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Gemma 3 27b It | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Gpt Oss 20b | 45.0 | 0 | 0.0% | 400ms | 1000ms | |
| Llama 3.2 1B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 8B | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Qwen3 Coder 480B A35B Instruct | 45.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| huggingface/deepseek-ai/DeepSeek-V3-0324- | 34.0 | 1 | 0.0% | 457ms | 457ms | |
| Qwen: Qwen3 Coder 480B A35B (free) | 15.0 | 4 | 0.0% | 9932ms | 15554ms | 🔧 |
| deepseek/deepseek-v3.1 | 15.0 | 3 | 0.0% | 17892ms | 37687ms |
How to Use Huggingface on Antbase
All Huggingface models are available through Antbase's unified OpenAI-compatible API. You can either route directly to a specific Huggingface model or let ANT's intelligent router automatically select the best model for your task.
python
from openai import OpenAI
client = OpenAI(
base_url="https://antbase.ai/v1",
api_key="your-ant-api-key",
)
# Let ANT auto-route (may use Huggingface if optimal)
response = client.chat.completions.create(
model="ant:auto",
messages=[{"role": "user", "content": "Hello!"}],
)Why Use Huggingface Through Antbase?
- •No separate account needed — access Huggingface models instantly through your Antbase API key.
- •Intelligent routing — ANT automatically selects the best Huggingface model for each task, or falls over to another provider if Huggingface is slow or down.
- •Same token prices — Antbase charges zero markup on Huggingface token prices. You pay exactly what Huggingface charges.
- •Automatic failover — if Huggingface has an outage, your requests seamlessly route to an equivalent model on another provider.