BNB's Pasteur Fork Shipped. The Price Didn't Move.
The BNB Smart Chain Pasteur hard fork activated at 02:30 UTC on August 25, 2026, with no chain halt and no rollback. It is the only hard protocol catalyst in the current two-week window, and BNB spent the week ahead of it trailing both Bitcoin and Ethereum.
This is analysis, not financial advice. Crypto assets are volatile and the majority of retail speculative accounts lose money over a full cycle; treat any single-asset thesis accordingly.
What a skeptic says first
The catalyst has already fired. Pasteur went live before this post did. Whatever the market intended to price in around the upgrade had a window to do it before you read this. A post-hoc protocol post is a "what changed" piece, not a call on an event ahead.
The headline throughput numbers are testnet, not mainnet. The 1,237 to 2,324 transactions-per-second figure comes from QANet, an internal controlled environment. Average gas per block rose from 46.35 million to 84.15 million against a 100 million limit under the same conditions. BNB Chain itself flagged that these do not represent confirmed mainnet performance. Anyone quoting "88% faster" as a live result is quoting a lab.
This is an exchange-affiliated token, and that history is not ancient. Binance Holdings and founder Changpeng Zhao pleaded guilty to Bank Secrecy Act violations in November 2023, producing $4.3 billion in combined penalties and a four-month federal sentence for Zhao. The CFTC settlement ordered $1.35 billion in disgorgement plus a $1.35 billion civil penalty. The SEC action was dismissed with prejudice on May 29, 2025, so no active enforcement matter stands today — but an independent compliance monitor remains in place for five years from the DOJ settlement, running into late 2028. Exchange-token risk here is concentration and counterparty risk, not a pending case.
The regime is hot, which raises entry risk rather than lowering it. Bitcoin ran roughly 22% in seven days to a $79,461 intraday high, its strongest week of 2026. Ethereum added about 31% over the same stretch. Publishing an entry into a vertical week means the drawdown risk on that entry is elevated, not reduced. CoinShares' James Butterfill called the move "a macro story, not a crypto one" — softer inflation and weaker payrolls, not anything endogenous to the asset.
A supply figure that is widely misreported
Several tokenomics aggregators show BNB as "66.58% unlocked" with roughly 66.8 million tokens "locked." That reading is wrong, and it matters if you are sizing dilution risk.
BNB launched with a 200,000,000 genesis supply. Quarterly Auto-Burn plus BEP-95 real-time burns have permanently destroyed about 66.83 million of them. Those tokens are burned, not locked. They are not coming back and there is no vesting schedule behind them — BNB's vesting ended in 2021.
Against current total supply, circulating supply is 99.99%. There is no unlock overhang to underwrite. The next scheduled supply event is the October 2026 quarterly burn, which removes supply rather than adding it.
BNB Supply and Burn Table
| Metric | Value | Source |
|---|---|---|
| Genesis supply | 200,000,000 BNB | BNB Chain |
| Total supply after 36th burn | 133,166,127.91 BNB | BNB Chain, 16 Jul 2026 |
| Circulating supply | 133,155,707 BNB (99.99% of total) | Kraken, 25 Aug 2026 |
| Cumulative burned | ~66,833,872 BNB | Derived from genesis less total |
| 36th quarterly Auto-Burn | 1,615,827.795 BNB (~$931.7M) | BNB Chain, 16 Jul 2026 |
| Remaining vesting unlocks | None — schedule ended 2021 | Tokenomist |
| Long-term burn floor | 100,000,000 BNB | BNB Chain Auto-Burn program |
The data
Every return below is measured over the same window and pulled on August 25, 2026. An unbenchmarked number is not a number.
Benchmark Table
| Asset | Price (25 Aug 2026) | 7-Day | 30-Day | All-Time High | Distance To ATH |
|---|---|---|---|---|---|
| BNB | $704.73 | +16.0% | +22.2% | $1,369.99 (13 Oct 2025) | +94.4% |
| Bitcoin | $79,106.77 | +23.7% | +19.5% | $126,198.00 (6 Oct 2025) | +59.5% |
| Ethereum | $2,507.22 | +31.4% | +31.2% | $4,946.05 (25 Aug 2025) | +97.3% |
All-time highs per CoinGecko, the same source the Hot Coins Tracker uses. Prices and distances verified 25 August 2026.
What Pasteur actually shipped| Proposal | What it changes | Why it matters |
|---|---|---|
| BEP-682 | Removes duplicate validator entries in bridge signature verification | The signature threshold now counts distinct validators. Duplicate counting is a known bridge-exploit vector |
| BEP-695 | Strips authority from rotated validator keys; extends blacklist protection to signature-based governance votes | Closes the gap where a retired key still carried weight in staking and governance |
| BEP-675 | Moves block execution from validators to builders; builders submit pre-executed blocks | The throughput change. Blocks fill further before the validator deadline. Block time is unchanged at 450ms |
The thesis
Two of the three proposals in Pasteur are bridge and validator security fixes. That is the part worth paying attention to, and it is the part nobody writes headlines about.
Cross-chain bridges have been the single most expensive failure surface in this industry. BEP-682 closes a specific class of that failure — a validator set where the same signer can be counted more than once against a threshold. BEP-695 removes standing authority from rotated keys. Neither generates a price move on the day. Both reduce the probability of the kind of event that removes nine figures from a chain's TVL in an afternoon.
The third, BEP-675, is the throughput item, and it is structural rather than parametric: block time stays at 450 milliseconds, but separating execution from validation lets builders hand over blocks that are already executed. Blocks fill further inside the same window. That is a real change to capacity, and it is also the change most likely to be overstated, because the only numbers available are from a controlled test environment.
Set against that: BNB rose 16% in the week the market rose 22% to 31%. It underperformed both majors into its own catalyst. Read charitably, the upgrade is not in the price. Read uncharitably, the market looked at a security-and-plumbing release and correctly concluded it does not change the earnings profile of an exchange-affiliated token.
What would falsify this. Three things, in order of how fast they would show up:
Mainnet throughput does not follow the testnet result. If sustained mainnet TPS and average gas per block do not move materially above pre-fork levels within 30 days, BEP-675 is a lab number and the capacity argument is dead. This is publicly measurable — no interpretation required.
A bridge or validator incident post-fork. An exploit that routes through the surfaces BEP-682 and BEP-695 were meant to harden would invalidate the security case outright.
BNB continues to lag the majors through the next 60 days. Persistent underperformance through a favourable tape stops being a discount and starts being a repricing.
The macro backdrop is not supportive of precision here. The rally that carried this week is a Fed-expectations story, with Jackson Hole guidance and Core PCE both landing on August 26. A macro-driven tape can undo an asset-specific thesis in a session.
The track record
Hot Coins has run since 2017. The record is published in full — including the picks that went to zero — on the CryptoBull Hot Coins Tracker, which now carries 51 picks with an average all-time-high return of +1,499% from the original call date.
The headline entries: Chainlink at +4,796%, NANO at +4,603%, Helium at +3,990%, all measured from the date of the call. The tracker also carries delisted rows and a token-swap row that reports "n/a" rather than a flattering percentage, because a record that only shows the winners is not a record.
BNB enters that table at the low end. Its all-time high of $1,369.99 was set in October 2025 — before this call — so the +94.4% figure in the tracker is distance to a prior peak, not an achieved return. That is a materially different thing from LINK's number, and it is worth saying plainly rather than letting the column imply otherwise.
Last week's pick was XRP into the August 19 White House meeting — a narrative catalyst, explicitly labelled as one. This week's is a protocol change with a confirmed on-chain result. The difference is the point.
This is analysis and not financial advice. CryptoBull is independent and takes no payment from projects covered. Digital assets are volatile and most retail speculative positions lose money over a full market cycle. Prices, supply figures and all-time highs in this post were verified on August 25, 2026 and will drift. Do your own research and size positions you can afford to lose entirely.