Qwen3 Next 80B A3B Instruct
qwen/qwen3-next-80b-a3b-instruct
Qwen3-Next uses a highly sparse MoE design: 80B total parameters, but only ~3B activated per inference step. Experiments show that, with global load balancing, increasing total expert parameters while keeping activated experts fixed steadily reduces training loss.Compared to Qwen3’s MoE (128 total experts, 8 routed), Qwen3-Next expands to 512 total experts, combining 10 routed experts + 1 shared expert — maximizing resource usage without hurting performance.
The Qwen3-Next-80B-A3B-Instruct performs comparably to our flagship model Qwen3-235B-A22B-Instruct-2507, and shows clear advantages in tasks requiring ultra-long context (up to 256K tokens).
Fonctionnalités
API sans serveur
Documentation
qwen/qwen3-next-80b-a3b-instruct is available via Novita's serverless API, where you pay per token. There are several ways to call the API, including OpenAI-compatible endpoints with exceptional reasoning performance.
Déploiements à la demande
Documentation
On-demand deployments allow you to use qwen/qwen3-next-80b-a3b-instruct on dedicated GPUs with high-performance serving stack with high reliability and no rate limits.
Sans serveur disponible
Exécutez des requêtes immédiatement, ne payez que pour l’utilisation
Entrée$0.15 / M Tokens
Sortie$1.5 / M Tokens
Utilisez les exemples de code suivants pour intégrer notre API :
1from openai import OpenAI
2
3client = OpenAI(
4 api_key="<Your API Key>",
5 base_url="https://api.novita.ai/openai"
6)
7
8response = client.chat.completions.create(
9 model="qwen/qwen3-next-80b-a3b-instruct",
10 messages=[
11 {"role": "system", "content": "You are a helpful assistant."},
12 {"role": "user", "content": "Hello, how are you?"}
13 ],
14 max_tokens=32768,
15 temperature=0.7
16)
17
18print(response.choices[0].message.content)Infos
Fournisseur
Qwen
Quantification
bf16
Fonctionnalités prises en charge
Longueur du contexte
131072
Sortie maximale
32768
Serverless
Pris en charge
Function Calling
Pris en charge
Structured Output
Pris en charge
Reasoning
Pris en charge
API Anthropic
Pris en charge
Capacités d’entrée
text
Capacités de sortie
text