Why Did DeepSeek Just Raise API Prices by Up to 1,100% — Right After China's Open-Weight AI Blitz?

In a five-day window, three of China's top AI labs made moves that look contradictory on the surface. DeepSeek raised API prices by as much as 1,100% on certain tiers. Alibaba, on the same week, open-weighted a 2.4-trillion-parameter flagship model it had never released before. Zhipu AI shipped GLM-5.3, boosting coding benchmarks by roughly 6x using the exact same base model as its predecessor — no retraining involved. Together, these three moves signal that China's AI labs are shifting from competing on price alone to competing on pricing power itself.

1. Three decision failures: headlines, cheapest-assumption, sleeping laptops

1. Treating "11x", "1,100%", and "350%" as the same hike: all three are accurate, but they describe different billing lines — cache-hit input, output, and cache-miss input. Mixing them mis-sizes the invoice by an order of magnitude.
2. Assuming the official DeepSeek API is still always cheapest: at peak hours, DeepSeek's own list price now sits above several resellers (GMI Cloud, Novita, and others currently list V4 Pro below the new official peak). "Chinese model = cheapest model" is no longer a safe default.
3. Running peak-sensitive agent evals on a laptop that sleeps: the announcement explicitly asks for more flexible workload scheduling. A closed lid during Beijing off-peak is how you buy the expensive hours by accident.

2. Timeline: what happened, and when

Zoom out further and the picture gets more interesting: on Jul 30, OpenAI cut prices on its cheapest tier (GPT-5.6 Luna, down 80%), then on Aug 6–7 made Luna the free default with unlimited text chats. In other words, while Chinese labs were raising prices and opening up flagship weights, US labs were cutting prices and going free at the consumer layer — at the exact same time. That's not a coincidence; it's two sides of the same pricing fight. Earlier product-level write-ups: Qwen3.8-Max launch, V4-Flash official benchmarks, and GPT-5.6 Luna's 80% cut.

3. The numbers: what actually changed

DeepSeek's price hike, tier by tier (effective Aug 17, 00:00 Beijing time; peak hours are 9am–12pm and 2pm–6pm Beijing time)

The headline "1,100%" figure everyone quoted applies specifically to peak-hour cache-hit input pricing — the tier that started closest to free. Output pricing, which matters more for most real-world bills, rose 350%. Independent cost modeling by third-party analysts found that a realistic heavy-usage workload (roughly 84M tokens/month, mostly off-peak, half cache hits) sees a bill increase closer to 1.8x — real, but far below the scariest headline numbers.

Qwen3.8-2.4T-A95B (Qwen3.8-Max open weights): key specs

GLM-5.3 vs GLM-5.2: same base model, post-training only

These are Zhipu's own reported numbers — no independent third-party re-run has been published yet. GLM-5.3 still trails GPT-5.6 Sol (34.6%) and Claude Fable 5 (33.7%) on Terminal-Bench 3.0; it's a top open-weight result, not an outright frontier win.

4. Breaking down the three strategies

DeepSeek: from flat-rate pricing to time-of-day pricing — this is a capacity problem, not a strategy pivot

The easiest misread of DeepSeek's move is "China's cheapest model finally caved to margin pressure." Look closer at the structure and it reads more like the opposite: a company making its compute constraints visible in the price sheet for the first time. DeepSeek's old flat, always-cheap pricing worked as a customer-acquisition tool as long as GPU capacity kept pace with demand. Once usage grew exponentially and capacity didn't, something had to become explicit — and "encouraging more flexible workload scheduling" in the official announcement is corporate-speak for "our peak-hour compute is now scarce, please shift your load yourself."

One detail international coverage mostly missed: at peak hours, DeepSeek's own official API price is now higher than several third-party resellers (GMI Cloud, Novita, and others currently list V4 Pro below DeepSeek's new peak rate). The assumption that "the official API is always the cheapest way to run DeepSeek" — a core part of its reputation — has been broken for the first time.

Alibaba: open weights buy ecosystem goodwill; a custom license protects the revenue ceiling

Qwen3.8-Max's open-weighting isn't a straightforward act of generosity. Alibaba did two things simultaneously: it published the full 2.4T-parameter checkpoint for free download, and it attached a custom license — not the permissive Apache 2.0 used for smaller Qwen models — that requires any "Model-as-a-Service" or "AI Work Assistant" business earning over $50 million in any 12-month period to negotiate a separate commercial license, and requires products with 100M+ monthly active users or $20M+ in monthly revenue to prominently display the model's name.

The logic: give away the weights to win developer mindshare (especially internationally, where "made-in-China model" still carries some hesitation among enterprise buyers), while keeping pricing leverage over the handful of companies actually capable of building a competing inference business on top of it. That's a materially different bet than Meta's Muse Glimmer, which ships under unrestricted Apache 2.0 — "open weights" doesn't mean the same thing across these two releases.

One rumor worth killing explicitly: claims circulated online that Alibaba's license bans downloads from the US, EU, UK, and South Korea. That's false. The published license text contains no geographic or territorial clause of any kind — a useful reminder that in a release cycle this fast, checking the primary source (the LICENSE file, not the announcement thread) takes seconds and saves you from repeating a debunked claim.

GLM-5.3: no new base model, just a bigger post-training bet — and that's the real story

The most interesting fact about GLM-5.3 isn't the score, it's the method: same 743B-parameter base as GLM-5.2, no retraining, and a roughly 6x jump on Terminal-Bench 3.0 (4.6% → 28.3%) purely from scaling up reinforcement learning environments in post-training. This confirms a trend that's been building industry-wide for months — as pretraining scaling laws show diminishing returns, post-training RL scale is becoming an independent performance lever with a much lower cost floor than retraining a new foundation model. That's a meaningfully lower barrier to entry, and it's why mid-tier labs without OpenAI-scale compute budgets can still close the gap on agentic and coding benchmarks.

5. Head-to-head: is DeepSeek still the cheapest frontier-class model?

RMB-to-USD conversion at ~¥7.15/$1, approximate. The short answer: no. Even after accounting for the hike, DeepSeek V4-Pro's off-peak rate is still well below Claude Opus 5, but it's no longer the outright cheapest option on the table — both Qwen3.8-Max's international pricing and OpenAI's Luna now undercut DeepSeek's off-peak rate. "Chinese model = cheapest model" was true for most of 2025 and early 2026; it isn't a safe assumption anymore.

6. What's disputed or unverified

• The "1,100%" headline is technically accurate but misleading without context. It applies only to peak-hour cache-hit input pricing, the tier that started nearest to zero. Output pricing — the cost that dominates most real bills — rose 350%. Different outlets have quoted different tiers as if they were the whole story.
• Claims that Qwen3.8-Max runs on Alibaba's in-house Zhenwu M890 chips (reported by several Chinese financial outlets as evidence of a fully domestic-silicon inference stack) have not been independently confirmed by Alibaba's own technical documentation or third-party benchmarks. Treat this as vendor-adjacent, unverified reporting until confirmed.
• GLM-5.3's reported discovery of a "serious vulnerability" in Cursor comes from VentureBeat's reporting and Zhipu's own disclosure; specific technical details of the vulnerability have not been made public, so the claim should be read as a vendor-sourced, not independently audited, security finding.
• Reports that China's Ministry of Commerce may be preparing retaliatory export controls on AI/semiconductor technology are speculative and sourced to unconfirmed media reports, not an official announcement. Treat as background context, not established fact.

7. Why this matters: two price wars running in parallel

Place this in the bigger frame and a pattern emerges. Over roughly the past month, China's top labs have shipped major releases at a pace domestic financial media has started calling "three model updates a week" (一周三更) — DeepSeek, Alibaba, and Zhipu, plus Moonshot's Kimi K3 (open-weighted Jul 16, 2.8T parameters) and MiniMax H3 before them. Chinese coverage broadly frames this as Chinese open-weight releases "forcing a global repricing of the AI industry" — a framing that's more assertive than most English-language coverage of the same events.

Meanwhile, US labs are running the opposite play at the consumer layer: OpenAI cut prices 80% on its cheapest tier (Jul 30) then made that model free and unlimited for all users a week later (Aug 6–7); Google shipped a coding-focused model at half the price of its three-week-old predecessor (Aug 13). So while Chinese labs open-weight flagships and introduce tiered, higher pricing on the compute-constrained top end, US labs are racing toward free and cheap at the consumer end. Both are real strategies; they're just optimizing for different parts of the funnel.

There's also a geopolitical layer worth naming carefully. Moonshot's Kimi K3 open-weighting in July already drew US security scrutiny; Alibaba choosing this specific window to open-weight a 2.4T flagship has been read by some analysts as a move to lock in international mindshare and a "technological parity" narrative before any potential regulatory tightening. That's an informed interpretation, not a confirmed fact — but it's part of the context that's hard to see if you're only reading English-language tech press, which has largely covered these releases as isolated product news rather than as a coordinated national pattern.

8. Five steps: recalculate the bill and reroute

1. Separate the headline tiers: 1,100% is V4-Pro peak cache-hit input. Output is 350%. Cache-miss input is 200%. Match the official line item before you change routing.
2. Slice the last 30 days by Beijing hour: peak is 9am–12pm and 2pm–6pm. Your peak share decides whether you see ~1.8x or the headline multiple.
3. Estimate cache-hit rate: a heavy, mostly off-peak, half-hit workload was modeled at about 1.8x. Peak plus low hit rate is where headlines become real.
4. Read the Qwen license before you ship: personal and internal use is fine. MaaS / AI Work Assistant over $50M in any 12-month window needs a separate commercial license. 100M+ MAU or $20M+ monthly revenue must display the model name. No geographic ban exists.
5. Put deferrable work on an always-on remote Mac: batch jobs and overnight agent evals should miss Beijing peak. A sleeping laptop cannot hold that window.

9. Off-peak eval host decision matrix

10. FAQ

Is DeepSeek still cheaper than GPT-5.6 or Claude after the price hike?
Its off-peak rate is still cheaper than Claude Opus 5, but it's no longer the single cheapest option overall — OpenAI's GPT-5.6 Luna ($0.20/$1.20 per million tokens) and Alibaba's international Qwen3.8-Max pricing ($2/$6) now undercut DeepSeek's new off-peak rates on at least one dimension. DeepSeek is still relatively cheap for a frontier-class model, just not the outright cheapest anymore.

Can I use Alibaba's Qwen3.8-Max open weights for free in a commercial product?
Yes, for most use cases — personal projects and internal enterprise use are unaffected. The catch applies only if you're running a "Model-as-a-Service" or "AI Work Assistant" business that has earned over $50 million in any consecutive 12-month period; that tier requires a separate commercial license from Alibaba.

Is Qwen3.8-Max banned or restricted for US, EU, or UK users?
No. That claim circulated online but is false — the published license contains no geographic restriction of any kind. The restrictions are revenue-based (tied to how much money your service makes), not tied to where you or your users are located.

What's actually different between GLM-5.3 and GLM-5.2?
Nothing at the base-model level — both use the same 743-billion-parameter foundation model. The performance gains (roughly 6x on Terminal-Bench 3.0) come entirely from scaling up reinforcement learning during post-training, with no retraining of the base model.

Will Meta actually open-source its flagship model, not just the smaller Muse Glimmer?
Not yet. Muse Glimmer is a 30B distilled model, not Meta's real flagship. CEO Mark Zuckerberg has said open weights for the larger, closed Muse Spark 1.2 are coming "soon," which — if it happens — would make it the first US flagship-tier model released openly. As of this writing, that release hasn't happened; treat it as a stated intention, not a confirmed fact.

Sources: DeepSeek's official pricing announcement, cross-checked against Wall Street CN, IT Home, AIGC.cn, and V2EX community discussion; Alibaba's official Qwen model repositories (Hugging Face / ModelScope) and South China Morning Post's reporting on license terms; Zhipu (Z.ai)'s official GLM-5.3 technical page, plus VentureBeat and StableLearn coverage; Meta AI Research's official blog and VentureBeat's coverage of Muse Glimmer; Chinese financial outlets (Yicai/第一财经, Sohu Finance) on the pacing and framing of China's open-weight release cycle. Pricing, license terms, and benchmark figures reflect publicly available information as of publication. Verify the latest official pricing and license terms before republishing, and note that details flagged above as unverified (domestic chip claims, the Cursor vulnerability report, and export-control rumors) have not been independently confirmed.

11. What the repricing is worth — and where a laptop still fails

The timeline, three data tables, and head-to-head rates are enough to make one decision: split the headline by billing line, then recompute against your Beijing peak share and cache-hit rate. Open weights are not a free-for-all until you have read the license.

The limits are equally clear. GLM-5.3 scores are vendor-reported. Zhenwu M890, the Cursor vulnerability claim, and export-control rumors remain unverified. FX is an estimate at ~¥7.15/$1. Coverage is not a substitute for your own traffic slice and license review.

If the next step is overnight batch evals, Beijing-scheduled agent runs, or trying Qwen open weights inside Cursor / OpenClaw, a sleeping laptop is how you miss the cheap hours. Put the scheduler on an always-on Apple Silicon remote Mac and sync the eval repo over SFTP/rsync. SFTPMAC remote Mac rental gives you native macOS, low-latency collaboration, and 24/7 uptime — a better place to turn this pricing shift into a reproducible routing change.