DeepSeek V4 Pro Released and Potentially Withdrawn Due to Wrong Version
DeepSeek quietly released the V4 Pro 0813 version around August 13. However, it faced criticism for marginal or even regressed performance in general benchmarks and was allegedly withdrawn by officials due to a 'wrong version' error. The model showed impressive results in cybersecurity and coding domains, achieving top-tier recall rates at a very low cost, but the stark contrast between its mediocre general performance and specialized breakthroughs sparked widespread controversy in the community.
Confirmed
- Marginal General Performance Improvement: Multiple developers pointed out that V4-0813 performed almost identically to V4-Flash-0731 in most benchmarks, with SciCode scores even regressing, and its Intelligence Index only 1 point higher than Flash. @haider1 added that while it beat Opus 4.8 at an extremely low cost, the score gap between frontier models is hard to perceive in daily use.
- Code and Text Rankings: According to Chatbot Arena's AutoEval, DeepSeek-V4-Pro (Max) scored 1607 in the WebDev Code Arena, ranking 2nd among open-source models, and scored 1465 in the Text Arena, ranking 5th.
- Outstanding Security but Specialized: Under pass@3 settings, V4 Pro rediscovered 87.5% of baseline CVEs, higher than Opus 5 and Qwen 3.8's 81.3%. @teortaxesTex tested it finding an RCE vulnerability in an open-source project in under 30 minutes but noted it ranked last in precision despite top recall. Tests also showed V4 Flash outperformed the Pro version in Canvas coding tasks.
- Cost and Volume Disadvantages: The cost per task is only $0.14, 17 times cheaper than the higher-ranked Kimi K3. However, @scaling01 noted that its parameter volume reaches 1.6T, 5 times larger than the Flash version, leading to accusations of extremely low cost-effectiveness.
- Suspected Wrong Version and Withdrawal: @orange and @op7418 indicated the version was likely withdrawn due to a 'wrong version'; while the API price page lists 0813, the update log does not mention it, and promotional banners were removed. @orange suggested that underwhelming performance might have led to the cancellation of a planned price hike.
Unconfirmed
- Score Gaming Controversy: @teortaxesTex mentioned the model's 11-point surge in the Vals index might be inflated, accused of relying solely on pass@2 gaming, but officials have not responded.
- Reason for Withdrawal: The withdrawal is currently only speculated as a 'wrong version'; there is no official notice confirming the specific cause.
Why it matters
The release and withdrawal of DeepSeek V4 Pro highlight the dilemma of 'marginal gains' in current general LLM benchmarks. However, its breakthrough performance and low cost in security and coding domains offer new ideas for vertical applications. The official's low-key promotion and withdrawal also reflect the complex game between frontier model iteration and evaluation expectations.
2026-08-13 ~ 2026-08-14 · 18 related posts
- Episode 1: DeepSeek V4 Pro Launch Imminent, API and Timeline Leaked(2026-08-11, 5 posts)
- Episode 2: DeepSeek V4 Specs and Pricing Leak Spark Debate(2026-08-11, 5 posts)
- Episode 3: DeepSeek launches V4-Pro with peak pricing, API price hike sparks debate(2026-08-12, 62 posts)
- Episode 4: DeepSeek V4 Pro Released and Potentially Withdrawn Due to Wrong Version(2026-08-13, 18 posts)
- Episode 5: DeepSeek Releases V4 Pro(2026-08-13, 5 posts)
- Episode 6: DeepSeek Hikes V4 API Prices Up to 1100% Against the Trend, Adds Peak/Off-Peak Billing(2026-08-15, 5 posts)
- Episode 7: DeepSeek Hikes V4 API Prices Up to 1100%, Introduces Peak/Off-Peak Pricing(2026-08-17, 5 posts)
Primary sources
- DeepSeek v4-pro Beats Opus at Minimal Cost, But Real-World Differences Remain Marginal — haider1 · 2026-08-13
- DeepSeek V4-Pro Ranks #2 Open-Weight Model, Accused of Relying on pass@2 — teortaxesTex · 2026-08-13
- DeepSeek Cybersecurity Test: Top Recall but Bottom Precision — teortaxesTex · 2026-08-13
- DeepSeek V4 Pro Finds RCE Vulnerability in Open-Source Project in Under 30 Minutes — teortaxesTex · 2026-08-13
- DeepSeek V4-0813 Shows No Gains Over July Version in Benchmarks — teortaxesTex · 2026-08-13
- DeepSeek V4-0813 Reported as Virtually Identical to Flash Version — teortaxesTex · 2026-08-13
- DeepSeek-V4-Pro Ranks #2 Among Open Models in Code Arena — arena · 2026-08-13
- DeepSeek-V4-Pro Ranks #5 in Open Text Arena Category — arena · 2026-08-13
- DeepSeek V4 Pro Mocked for Scoring Only 1 Point Higher Than Flash — ns123abc · 2026-08-13
- [source] DeepSeek V4 Pro Tops Security Benchmark with 87.5% Vulnerability Discovery Rate — solyarisoftware · 2026-08-13
- Surprising Benchmark: DeepSeek V4 Flash Outperforms Pro in Coding — solyarisoftware · 2026-08-13
- DeepSeek V4 Pro Quietly Released: Intelligence Index 53, Above-Average Pricing — ns123abc · 2026-08-13
- [source] DeepSeek V4 Pro Update Suspectedly Pulled Amid Abnormal Benchmark Scores — op7418 · 2026-08-13
- Rumor: DeepSeek v4 Pro Benchmark Underwhelms, Price Hike Likely Canceled — oran_ge · 2026-08-13
- [source] DeepSeek V4 Pro Taken Down Due to Accidental Wrong Version Release — oran_ge · 2026-08-13
- DeepSeek V4 Pro Underwhelms: 5x Larger But Only 1 Point Higher — scaling01 · 2026-08-13
- Testing 1.6T Param DeepSeek V4 Pro: Average Coding, Outpriced by GPT-5.6 — curiousily_ · 2026-08-13
- DeepSeek V4 Pro 0813 fails same demo twice while Flash succeeds, user reports — neverontime5 · 2026-08-14