Open Source
Quasar 438B开源
Massive Mixture-of-Experts open-weight model with 38B active parameters per token.
Intelligence
0.0 #700
Top 100% tier
Output speed
95 t/s
TTFT ~0.8s · ITL 10.5 ms/tok
Blended price (3:1)
$0.7 /1M
In $0.4/M · Out $1.6/M
Context
128K
≈ 328 A4 pages · max out 32768
Quality Domain Radar
6-Axis Frontier Evaluation vs Category Average & Flagship
Quasar 438B
Claude Mythos Preview (#1)
Average (68)
Radar axes are approximated from category scores where direct benchmark measurements are missing — see 评分方法.
API deployment
Compare hosts & latency
Throughput, TTFT and cache pricing across Groq, Cerebras, Together, DeepInfra, Fireworks, Bedrock and Azure.
GroqCerebrasTogether AIDeepInfraFireworks AIAWS BedrockAzure AI
Groq
Cerebras
Together AI
DeepInfra
Fireworks AI
AWS Bedrock
Azure AI各榜单得分
分类能力得分
基准测试分项
模型详细参数
发布日期
2026年8月1日
参数量
438B (38B active)
开源协议
Open Weights
上下文窗口
131,072 tokens
输入价格
$0.4/M
输出价格
$1.6/M
输出速度
95 t/s
首Token延迟
—
Estimated monthly bill
Estimated monthly spend
$0.00
Calculating savings vs GPT-4o...
Cite & embed
GitHub Markdown badge
[](https://llmpodium.com/models/quasar-438b)BibTeX
@misc{llmpodium2026quasar438b,
title = {Benchmark Analysis and Intelligence Evaluation of Quasar 438B},
author = {LLMPodium Research},
year = {2026},
url = {https://llmpodium.com/models/quasar-438b}
}替代模型推荐
基于能力、定价和基准测试得分与 Quasar 438B 最接近的替代模型。
