Gemini 4 Leaks Flood Out as Bloomberg Reports Internal Struggles
From October 1-2, a wave of leaks around Google's rumored Gemini 4 / Gemini 4 Argon pointed to strong benchmark results, but Bloomberg reported the same days that Google insiders found real-world performance falling short of scores, especially in coding, and that the model is very large with high running costs that could pressure margins. The community erupted into a debate over "benchmaxxing vs. real capability." None of the information is official.
Confirmed
- Bloomberg reported: Alphabet is preparing to launch Gemini 4; the model performs well on industry benchmarks, but multiple insiders with direct access said employee real-world usage fell short of expectations, particularly struggling on some coding tasks. Internally, opinions are split on whether it has caught up with OpenAI and Anthropic. Sources spoke anonymously.
- Bloomberg further disclosed (relayed by firstadopter): Gemini 4 is reportedly a very large model with costly inference that could pressure Google's margins; experts also noted Google's tendency toward benchmark optimization.
- Google employee Kuba tweeted proudly about working on Gemini; developer sunjiao123sun quoted him, saying that without Kuba, Gemini 4 Argon would not have scored 91.9 on VibeCodeBench.
- LuminaBench posted that Gemini 4 Argon scored 53 on the AA Intelligence Index, tied with GPT-6 Astra, sitting near the Pareto frontier on the cost curve.
Unconfirmed
- A Reddit leak claimed Gemini 4 Argon topped 12 of 18 benchmarks against Fable 5.1, Opus 5.5 and GPT-6 Astra, including 19.6% on autonomous legal work; authenticity unverified.
- Blogger haider claimed Gemini 4 reaches SOTA on knowledge tasks with coding roughly at Astra/Fable 5.1 level — an unverified leak.
- teortaxesTex speculated from pricing that Argon may be a mid-tier model targeting Claude Sonnet (internal Sol tier).
- DrSingularity claimed Gemini 4 Argon "has launched," with no official or third-party confirmation.
Why it matters
- The contrast between Bloomberg's insider account and the flood of benchmark rumors made "do benchmarks reflect real capability" the core debate. Users like EstablishmentFun3205 questioned whether this is real progress or benchmaxxing; Angaisb and ishuagra02 cited past gaps between Google's official scores and actual performance for Gemini 3 / 3.1 Pro; koltregaskes and Neurogence urged waiting for public availability before judging; ChrisGPT called it a step in the right direction despite struggles.
- Cost is another key dimension: Bloomberg's report of a very large model pressuring margins versus LuminaBench's near-Pareto-frontier claim together point to the tradeoff between flagship capability and inference cost. If the benchmark data holds, Gemini 4 Argon could reshape the frontier-model landscape, and the tiering rumors suggest Google is building a multi-tier product line for the mid-to-high-end market.
2026-10-01 ~ 2026-10-02 · 35 related posts
- Episode 1: Suspected Gemini 4 Pro Spotted in Anonymous Arena Testing(2026-09-26, 8 posts)
- Episode 2: Google's Gemini 4 Enters Final Training Push, May Launch by Year-End(2026-09-26, 2 posts)
- Episode 3: Leaked Gemini 4 Pro Benchmark Sparks Buzz Ahead of OpenAI DevDay(2026-09-28, 4 posts)
- Episode 4: Gemini 4.0 Pro and Multiple New Models Tipped for Release Within Weeks(2026-10-01, 2 posts)
- Episode 5: Google Unveils Gemini 4 Argon with 1M-Token Output, Defense-First Rollout(2026-10-01, 118 posts)
- Episode 6: Gemini 4 Leaks Flood Out as Bloomberg Reports Internal Struggles(2026-10-01, 35 posts)
- Episode 7: Gemini 4 Argon Shines in Long-Horizon Coding Benchmarks(2026-10-01, 2 posts)
- Episode 8: Viral "Gemini 4 Argon" blog page turns out to be a hoax(2026-10-01, 6 posts)
- Episode 9: Google's Gemini 4 Argon agents free over 300 TiB of datacenter memory(2026-10-01, 3 posts)
- Episode 10: Reported Gemini 4 Argon Beats Quantum Algorithm Baseline by 40% in Minutes(2026-10-01, 2 posts)
- Episode 11: Google Engineer Praises Gemini 4 Argon's Troubleshooting Skills(2026-10-01, 3 posts)
- Episode 12: Polymarket Bets Surge on Google as Gemini 4 Rumors Swirl(2026-10-01, 2 posts)
- Episode 13: Gemini 4 Expectations Split as Google Restart Pretraining(2026-10-01, 2 posts)
- Episode 14: Blogger Defends Google's TPU Edge as Gemini 4 Hype Builds(2026-10-02, 2 posts)
- Episode 15: Rumors Claim Google Has Far Stronger Internal Models(2026-10-02, 2 posts)
- Episode 16: Reported Google Gemini 4 Argon supports 1M-token outputs(2026-10-02, 2 posts)
- Episode 17: Gemini 4 Praised as Astra-Class at One-Fifth the Price(2026-10-03, 3 posts)
- Episode 18: Gemini 4 Argon Tops Multiple Leaderboards, Sparking Debate(2026-10-04, 3 posts)
- Episode 19: Gemini 4 Argon Surfaces in Docs and Cloud Quotas Ahead of Launch(2026-10-04, 3 posts)
Primary sources
- Leaked Gemini 4 Argon benchmarks reportedly top 12 of 18 tests vs rivals — 141_1337 · 2026-10-01
- [source] Bloomberg: Gemini 4 aces benchmarks but struggles at real coding tasks internally — firstadopter · 2026-10-01
- [source] Rumor: Gemini 4 Argon scores 53 on AA Intelligence Index, matching GPT-6 Astra — LuminaBench · 2026-10-01
- Skeptic pushes back on 'Google is so back' hype, citing unreliable benchmarks — Angaisb_ · 2026-10-01
- "Google is so back" skepticism: model not even usable, benchmarks historically unreliable — Angaisb_ · 2026-10-01
- Gemini 4 Argon takes the lead on long-horizon software engineering benchmarks — ColbyHawker · 2026-10-01
- Gemini 4 reportedly struggling on key areas like coding, per Bloomberg — ChrisGPT · 2026-10-01
- Gemini 4 Argon reportedly live as Google's model cadence accelerates, unconfirmed — Dr_Singularity · 2026-10-01
- Gemini 4 Scores Badly on cua-bench, Fueling Benchmaxxing Concerns — burny_tech · 2026-10-01
- Influencers hype Gemini 4 Argon: GOATED or hopelessly benchmaxed? — thatroblennon · 2026-10-01
- Rumor: Gemini 4 spotted with SOTA knowledge scores, coding near Astra/Fable 5.1 level — haider1 · 2026-10-01
- Google grapples with internal skepticism over Gemini 4 coding performance ahead of launch — matt_slotnick · 2026-10-01
- Google Staff Trash-Talk Gemini 4 to Press, Colleague Fires Back — josh_wills · 2026-10-01
- Citing Gemini 3.1's benchmark claims, users urge skepticism toward Google's Gemini 4 Argon hype — ishuagra02 · 2026-10-01
- Google models ace benchmarks but feel mid in use — will Gemini 4 Argon differ? — VraserX · 2026-10-01
- Developers claim unreleased Gemini 4 Argon scores 91.9 atop VibeCodeBench — sunjiao123sun_ · 2026-10-01
- Speculation: rumored Gemini 4 'Argon' tier looks like Google's Sonnet-class rival — teortaxesTex · 2026-10-01
- Leaked internal review: Google employees unsatisfied with Gemini 4 on some coding tasks — sandeepyadavhas · 2026-10-01
- Don't hype Gemini 4 yet: benchmarks don't tell the full story, blogger warns — koltregaskes · 2026-10-01
- Did Google actually cook, or is this peak benchmaxxing? Reddit debates Gemini scores — EstablishmentFun3205 · 2026-10-01
- Did Google actually cook or is it just peak benchmaxxing? Reddit debates new model scores — EstablishmentFun3205 · 2026-10-01
- Gemini 4 Argon jumps to 57.6% on Terminal-Bench-Science but underperforms on TB 4.0 — JJitsev · 2026-10-01
- Gemini 4 Argon takes #1 on Arena's text leaderboard — Successful_Row_3209 · 2026-10-01
- Gemini 4 Argon review: the new flagship is just an incremental upgrade — DavidLinthicum · 2026-10-01
- Bloomberg: Gemini 4 is a very large model, appears affected by benchmaxxing — firstadopter · 2026-10-02
- DeepMind Senior Staff Engineer Calls Bloomberg's Gemini 4 Employee Gripe Report "BS" — MrZiruiWang · 2026-10-02
9 near-duplicate retellings: firstadopter · Neurogence · ChrisGPT · QuintinPope5 · kimmonismus · socialwithaayan · kimmonismus · daniel_mac8 · xennygrimmato_