SCANNING● LIVE DATAUTC 2026-06-29 00:00:00
VENDORS 23RECEIPTS 32CRITICAL 57D 0 NEWv0.9.4-rc1
gotnerfedgotnerfed
gotnerfed · index of recordfiled · · receipt #replit-2025-09-agent3-cost-overrun
RECEIPT#REPL-2025-09-Ocritical
Replit·Cloud AI coding workspace·scored21nerf/ 100

Replit Agent 3 launch triggers $1,000/week bills - users report 5-20x monthly cost spikes

Replit launched Agent 3 on September 10, 2025. Within a week, users reported single-week bills hitting $1,000 against previous monthly bills of $180-200. The Register and InfoWorld both covered the backlash.

Filed
2025-09-10
Observed
2025-09-10
Response window
UNCONTESTED · DAY 30
Jurisdiction
MIT · public
REPLITNERFED21SEP 10, 2025INDEX · OF · RECORD
§ 01THE CHANGEBEFORE → AFTER · CLOUD AI CODING WORKSPACE
− BEFORE · REMOVED2025-09-10
Typical Replit Agent user spend: ~$180-200/mo for sustained use
AGENT 2 + FLAT CHECKPOINT BILLING·baseline through 2025-06
+ AFTER · CURRENT2025-09-10
Same workload under Agent 3 + effort billing: $1,000+/week reported
AGENT 3 + EFFORT-METERED CHECKPOINTS·Agent 3 launched 2025-09-10
WAS~$180-200 / mo for typical sustained use
NOW
$1,000+ / week reported for same workload
5-20xREPORTED COST OVERRUN
PREDICTED IMPACT · EST.LOW CONFIDENCE
USER IMPACTSmall subsetof users on edge configurations
12-MO FINANCIAL DAMAGENegligible direct costsoft friction only
CHURN PRESSURE · 30Delevatedusers disable Agent 3 / cap budgets / migrate to Cursor or Claude Code
Estimate based on the receipt's Nerf Index (21 / 100), the severity flag, and the kind of change (other). How we estimate ↗
§ 02PANEL VERDICT3 OF 3 HARMFUL
PANEL RULING · INDEX OF RECORD
FOUNDHARMFUL★ UNANIMOUS
VOTES3 / 3·AVG79 / 100·SPREAD±6·RUBRICsealed
0182
0279
0376
01claude-haiku-4-5HARMFUL· sealed rubric
Launching a more sub-agent-heavy Agent on top of opaque effort billing produced order-of-magnitude bill increases that users could not predict; multiple outlets documented the pattern.
82/ 100
benignsevere
02gpt-5-nanoHARMFUL· sealed rubric
Vendor marketed Agent 3 as more cost-effective while user-observed bills increased 5-20x; the gap between claim and reality is the structural harm.
79/ 100
benignsevere
03gemini-2-5-flashHARMFUL· sealed rubric
Even granting more capability per task, the absence of pre-submission cost preview combined with multi-subagent fan-out is a textbook surprise-billing pattern.
76/ 100
benignsevere
Each model graded independently behind a sealed rubric. Verdicts are immutable and Brier-scored against subsequent reality - a public model-accuracy leaderboard, on us. Read the rubric ↗
§ 03VENDOR RESPONSE WINDOWCLOSED · UNCONTESTED
Replit · response window UNCONTESTEDUNCONTESTED
FILED2025-09-10
MARKED UNCONTESTED2025-10-10
NOW · DAY 292

Replit did not respond inside the 30-day window. This receipt now reads as uncontested in the public record.

§ 04EVIDENCE TRAIL3 SOURCES
CONTEXT · NARRATIVEfiled by gotnerfed · 2025-09-10

Agent 3 ran 'a lot more under the hood with subagents' for reviewing, planning, security checks, and fixes - each consuming an effort-metered checkpoint under the June 2025 billing model.

Agent 3 ran 'a lot more under the hood with subagents' for reviewing, planning, security checks, and fixes - each consuming an effort-metered checkpoint under the June 2025 billing model. One Register-quoted user reported: 'In the last week alone it charged me $1K since the new agent dropped whereas before it was never more than $180-200 a month for the same effort.' Another user said they 'blew through $70 in one night' at launch. Replit's own marketing called Agent 3 'our most advanced and autonomous Agent yet' and '3x faster and 10x more cost-effective than Computer Use models,' a claim users widely disputed. This event is the direct cause of the November 2025 forced billing migration already catalogued.

DETECTED2025-09-10KINDOTHERSEVERITYCRITICALSOURCES3VENDOR REPLYNONE
§ 05NEXT STEP · ESCAPE POD4 RANKED ALTS
◆ FREE / OPEN-SOURCE ALTERNATIVES4 ranked replacements for Replit.avg score 90 / 100
OpenCodeFREE / OSS
MIT·120k

Open-source Claude Code / Cursor replacement that lives in the terminal. Pick when you want zero vendor lock-in and any LLM backend.

MIT

Permissive (MIT) open-weight model targeting ChatGPT/Claude-class chat and agentic coding without API lock-in - the strongest open option for a self-hosted stack.

AGPL-3.0·60k

ChatGPT's Code Interpreter, but local and against any model.

Modified MIT

Near-frontier open-weight model to back a self-hosted ChatGPT/Claude replacement or an agentic coding stack.

List manually curated · last reviewed 2026-05-13Open full picker
Filed by gotnerfed · public open-source watchdog · MIT licensed · no vendor moneyINDEX · OF · RECORD