ModelScope
INFERENCE34 MODEL · CONTEXT MAKS 1M
Catatan dari sumber
Free API-Inference for registered users. Requires Alibaba Cloud account binding + real-name verification.
Setup dalam 5 menit
Pilih target snippet
OPENAI SDK (NODE.JS)
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api-inference.modelscope.cn/v1",
apiKey: process.env.MODELSCOPE_API_KEY,
});
const response = await client.chat.completions.create({
model: "glm-5-1",
messages: [{ role: "user", content: "Halo!" }],
});
console.log(response.choices[0].message.content);.ENV
MODELSCOPE_API_KEY=your-api-key-hereBelum punya API key? Ambil di halaman resmi ModelScope ↗.
Cara claim API key gratis
Langkah umum — detail pastinya ikuti halaman resmi ModelScope.
- 01Buka halaman API key ModelScope ↗
- 02Daftar akun baru, atau login kalau sudah punya.
- 03Generate API key di dashboard / settings.
- 04Pakai API key + Base URL
https://api-inference.modelscope.cn/v1di SDK atau HTTP client.
Model tersedia (34)
| MODEL | MODALITY | CONTEXT | OUTPUT | RATE LIMIT |
|---|---|---|---|---|
| GLM-5.1glm-5-1 | reasoning | 1.0M | — | — |
| MiniMax-M3minimax-m3 | vision + reasoning | 512K | — | — |
| Qwen/Qwen3.5-27B | Text | 256K | — | 2,000 RPD total; <=500 RPD/model (dynamic) |
| Qwen/Qwen3.5-35B-A3B | Text | 256K | — | 2,000 RPD total; <=500 RPD/model (dynamic) |
| GLM-4.7-FlashXglm-4-7-flashx | reasoning | 200K | 128K | — |
| deepseek-ai/DeepSeek-V4-Flashdeepseek-ai-deepseek-v4-flash | — | 8K | — | — |
| deepseek-ai/DeepSeek-V4-Prodeepseek-ai-deepseek-v4-pro | — | 8K | — | — |
| deepseek-ai/DeepSeek-V4-Pro-0813deepseek-ai-deepseek-v4-pro-0813 | — | 8K | — | — |
| deepseek-ai/DeepSeek-V4.1-Flashdeepseek-ai-deepseek-v4-1-flash | — | 8K | — | — |
| early-access/EA-29B-A4Bearly-access-ea-29b-a4b | — | 8K | — | — |
| MedAIBase/AntAngelMedmedaibase-antangelmed | — | 8K | — | — |
| meituan-longcat/LongCat-Flash-Litemeituan-longcat-longcat-flash-lite | — | 8K | 320K | — |
| MiniMax/MiniMax-M1-80kminimax-minimax-m1-80k | — | 8K | — | — |
| mistralai/Mistral-Large-Instruct-2407mistralai-mistral-large-instruct-2407 | — | 8K | — | — |
| MusePublic/Qwen-Image-Editmusepublic-qwen-image-edit | — | 8K | — | — |
| nex-agi/Nex-N2.5-mininex-agi-nex-n2-5-mini | — | 8K | — | — |
| nex-agi/Nex-N2.5-Pronex-agi-nex-n2-5-pro | — | 8K | — | — |
| opencompass/CompassJudger-1-32B-Instructopencompass-compassjudger-1-32b-instruct | — | 8K | — | — |
| OpenGVLab/InternVL3_5-241B-A28Bopengvlab-internvl3-5-241b-a28b | — | 8K | — | — |
| PaddlePaddle/ERNIE-4.5-0.3B-PTpaddlepaddle-ernie-4-5-0-3b-pt | — | 8K | — | — |
| PaddlePaddle/ERNIE-4.5-21B-A3B-PTpaddlepaddle-ernie-4-5-21b-a3b-pt | — | 8K | — | — |
| PaddlePaddle/ERNIE-4.5-300B-A47B-PTpaddlepaddle-ernie-4-5-300b-a47b-pt | — | 8K | — | — |
| PaddlePaddle/ERNIE-4.5-VL-28B-A3B-PTpaddlepaddle-ernie-4-5-vl-28b-a3b-pt | — | 8K | — | — |
| Qwen/Qwen3.5-122B-A10Bqwen-qwen3-5-122b-a10b | — | 8K | — | — |
| Qwen/Qwen3.5-397B-A17Bqwen-qwen3-5-397b-a17b | — | 8K | — | — |
| Qwen/Qwen3.8-27Bqwen-qwen3-8-27b | — | 8K | — | — |
| Qwen/Qwen3.8-Flash-Nextqwen-qwen3-8-flash-next | — | 8K | — | — |
| Shanghai_AI_Laboratory/Intern-S1shanghai-ai-laboratory-intern-s1 | — | 8K | — | — |
| Shanghai_AI_Laboratory/Intern-S1-minishanghai-ai-laboratory-intern-s1-mini | — | 8K | — | — |
| Shanghai_AI_Laboratory/Intern-S2-Previewshanghai-ai-laboratory-intern-s2-preview | — | 8K | — | — |
| stepfun-ai/Step-3.5-Flashstepfun-ai-step-3-5-flash | — | 8K | — | — |
| stepfun-ai/Step-3.7-Flashstepfun-ai-step-3-7-flash | — | 8K | — | — |
| XGenerationLab/XiYanSQL-QwenCoder-32B-2412xgenerationlab-xiyansql-qwencoder-32b-2412 | — | 8K | — | — |
| XGenerationLab/XiYanSQL-QwenCoder-32B-2504xgenerationlab-xiyansql-qwencoder-32b-2504 | — | 8K | — | — |
Model ini di provider lain
glm-5-1
minimax-m3
deepseek-ai-deepseek-v4-1-flash
FAQ
Apakah API ModelScope gratis?
Ya — menurut sumber: "Free API-Inference for registered users. Requires Alibaba Cloud account binding + real-name verification." Data dari mnfst/awesome-free-llm-apis, freellm.net, models.dev, terakhir di-sync 31 Okt 2026 — cek sumber untuk kondisi terbaru.
Ada berapa model gratis di ModelScope?
34 model tercantum per 31 Okt 2026, termasuk GLM-5.1, MiniMax-M3, Qwen/Qwen3.5-27B.
Berapa rate limit ModelScope?
Beda-beda per model, contoh: Qwen/Qwen3.5-27B: 2,000 RPD total; <=500 RPD/model (dynamic); Qwen/Qwen3.5-35B-A3B: 2,000 RPD total; <=500 RPD/model (dynamic). Limit lengkap per model, lihat tabel di atas.
Berapa context window maksimal di ModelScope?
1M.
DISINKRON 31 OKT 2026 · SUMBER: mnfst/awesome-free-llm-apis, freellm.net, models.dev