User ditches GPT-5.6-sol for DeepSeek V4 Pro: 'feels like Opus 7.0 landed'
ryunuck · reddit · 2026-09-03
A Reddit user recounts switching from GPT-5.6-sol to DeepSeek V4 Pro (0813), claiming it feels like Opus 6.0 or even 7.0 just landed—had they not known, they'd have believed it was the frontier's unreleased next-gen model.
Key observations:
- De-marketed language: code comments are 'inaccurate' not 'lying'; no wishy-washy hedging or psychosis-prevention lecturing—'working with an actual engineer.'
- Classic engineering vocabulary (e.g., 'expose declarations in the API') returns, which the author argues frontier models had trained out of users' critical thinking.
- Claims: bigger models show diminishing returns (100T won't beat 1T by much); the real moat is the training curriculum, and China is pulling ahead while US labs embed corporate agenda and impression management into alignment (citing Claude moderating its own subreddit).
Conclusion: don't believe the marketing or the hype threads—any off-the-shelf model can do the work.
More from Models
- Codex, Claude and Grok All Went Down at the Same Time — fekdaoui · 2026-09-03
- Gemini has improved exponentially, Grok slower and often unavailable: Damodaran — pdamodaran · 2026-09-03
- Hacker upgrades stolen account to $200 ChatGPT Pro with no fraud checks from OpenAI or bank — CucharitaDePalo · 2026-09-03
- Muse Spark 1.3 Update Gains Attention, Users Urged Not to Miss — D3VAUX · 2026-09-03
- 64GB Mac users debate best sub-40B MoE: is Qwen-3.8-35B worth the wait? — chibop1 · 2026-09-03
- Microsoft's MAI-Transcribe-2 hits 2.0% WER at 411x real-time, priced at $1.67 per 1,000 minutes — ArtificialAnlys · 2026-09-03