BenchmarkVerified

deepseek-chat ​

deepseek-chat ​ The deepseek-chat model has been upgraded to DeepSeek-V2-0628. Model's reasoning capabilities have improved, as shown in relevant benchmarks: Coding: HumanEval Pass@1 79.88% -> 84.76% Mathematics: MATH ACC@1 55.02% -> 71.02% Reasoning: BBH 78.56% -> 83.40% In the Arena-Hard evaluation, the win rate against GPT-4-0314 increased from 41.6% to 68.3%. The model's role-playing capabilities have significantly enhanced, allowing it to act as different characters as requested during conversations.

Evidence chain

DeepSeek API updatesOfficial · official

deepseek-chat ​ The deepseek-chat model has been upgraded to DeepSeek-V2-0628. Model's reasoning capabilities have improved, as shown in relevant benchmarks: Coding: HumanEval Pass@1 79.88% -> 84.76% Mathematics: MATH ACC@1 55.02% -> 71.02% Reasoning: BBH 78.56% -> 83.40% In the Arena-Hard evaluation, the win rate against GPT-4-0314 increased from 41.6% to 68.3%. The model's role-playing capabilities have significantly enhanced, allowing it to act as different characters as requested during conversations.

Open original source

External discussion

No reviewed external discussion has been linked yet. This does not change the evidence confidence.