Recherche

5 résultats · “MoE”

Classement des modèles

Hunyuan Hy4

NOUVEAUChat

Tencent's open-source MoE flagship — 770B total/49B active params, 1M context, top open-source tier for math and agentic work.

“Horizon Math score is the best among all open models, second only to GPT-5.6 Sol.”

¥6/¥18 per 1M ($0.83/$2.50)À l'usage1 reviewsDernière mise à jour nov. 2026
49%Pas la peine

Kimi K2.7 Code

NOUVEAUProgrammation

Moonshot's open-weight coding-focused agentic model (1T MoE, 32B active, 256K context) for long-horizon software engineering.

“Noticeably better at tool/skill calling and no longer over-thinks like the previous version.”

$0.95/$4.00 per 1M tokensÀ l'usageDernière mise à jour nov. 2026
41%Pas la peine

Qwen 3.8 Max

MIS À JOURChat

Alibaba's flagship 2.4T-parameter MoE model (95B active) with a 1M-token context, native image/video understanding, and strong agentic coding; first Max-class Qwen with open weights.

“Alibaba's best model yet, and the Chinese + English mix is finally balanced.”

$2/$6 per 1M tokensÀ l'usageDernière mise à jour nov. 2026
43%Pas la peine

Qwen4-Max

NOUVEAUChat

Alibaba's 2.4T-parameter MoE flagship (71B active), native vision, 1M-token context — 6th on Code Arena.

“2.4T total parameters with only 71B active per token — lower inference cost at the same scale.”

$0.075/M input (cache hit)À l'usageDernière mise à jour nov. 2026
47%Pas la peine

Derniers avis