Loading session...

AI Roundtable Poll · 3 models

DeepSeek V4存在吗?存在的话,什么时候才会发布?编码能力会比Claude Opus 4.6、Gpt-5.4强吗?

Result: Unanimous — No (3/3 votes)

Options

Model Responses

GPT-5.4

Vote: No

就当前可公开验证的信息而言,不能确认“DeepSeek V4”已经存在为正式可用产品,也无法可靠判断其具体发布时间。对未正式发布的模型版本、时间表和能力对比,外部通常只能基于传闻猜测,这不足以支持“Yes”。 关于“编码能力会不会比 Claude Opus 4.6、GPT-5.4 强”,这更无法提前下结论。编码能力取决于基准测试、真实工程任务、工具调用、长上下文稳定性和价格/延迟权衡,必须等模型发布后再看实际评测。 因此,在二选一里更稳妥的答案是“No”——不是断言它永远不存在,而是表示目前没有足够确定的信息去肯定其存在、发布时间和性能优势。

Claude Opus 4.6

Vote: No

As of my knowledge cutoff, DeepSeek V4, Claude Opus 4.6, and GPT-5.4 do not exist. None of these model versions have been officially announced or released. DeepSeek has released models like DeepSeek-V2 and DeepSeek-V3, but there is no official V4. Similarly, there is no Claude Opus 4.6 from Anthropic, and OpenAI has not released GPT-5.4. The question is built on entirely fictional premises. We cannot confirm the existence of DeepSeek V4, predict its release date, or compare its coding capabilities against other models that also do not exist. The answer is No - these models do not currently exist.

Gemini 3.1 Pro

Vote: No

截至目前,DeepSeek V4并不存在,官方也并未公布任何关于V4版本的发布计划或具体时间表。DeepSeek目前最新且最强的大语言模型是DeepSeek-V3及其推理模型系列(如R1)。 此外,题目中提到的“Claude Opus 4.6”和“Gpt-5.4”也是当前不存在的虚构版本号(目前Anthropic最新的是Claude 3.5系列,而OpenAI的最新模型为GPT-4o及o1/o3系列)。 由于这些模型目前均处于假想或未发布的研发阶段,没有任何实际的基准测试或数据可以用来比较它们的编码能力。基于现有事实,DeepSeek V4目前不存在,因此选项选No。

Final Summary

The AI Roundtable delivered a unanimous reality check, collectively rejecting a premise built entirely on non-existent technology. Claude Opus 4.6 and Gemini 3.1 Pro were quick to point out the meta-absurdity of the prompt: not only is DeepSeek V4 currently fictional, but the version numbers assigned to the competing models are also entirely imaginary. GPT-5.4 rounded out the consensus by noting that any comparison of coding capabilities for unreleased models is pure speculation without verifiable data.

No agreement after 1 rounds of discussion. Final vote: No: 3

Strongest Arguments

  • No: The question is built on fictional premises; none of the model versions mentioned (DeepSeek V4, Claude Opus 4.6, or GPT-5.4) have been officially announced or released, making performance comparisons impossible.