
Qwen4-Max vs Gemini 4: A 40% Cheaper Token That Buys Nothing Per Task
- OrcaNEWOrcaCyber: Hy4 Flash Abliterated / Uncensored2026-11-18$0.07 / $0.28 per 1M tokens · 180 tok/s
- tencentNEWTencent: Hy4 Flash2026-11-16$0.02 / $0.07 per 1M tokens · 223 tok/s
- GoogleNEWGoogle: Gemini 4 Carbon2026-11-10$2.00 / $10.00 per 1M tokens · 89 tok/s
- AnthropicNEWAnthropic: Claude Fable 5.22026-11-09$8.00 / $40.00 per 1M tokens · 98 tok/s
- orcaNEWOrca: OrcaC2 1.02026-11-079 tok/s
- OrcaOrcaCyber: GLM 5.3 Abliterated / Uncensored2026-11-04$3.00 / $5.00 per 1M tokens · 88 tok/s
- OrcaOrcaCyber: Hy4 Abliterated / Uncensored2026-11-04$0.80 / $1.60 per 1M tokens · 134 tok/s
- OrcaOrcaCyber: Kimi K3.1 Pro Abliterated / Uncensored2026-11-04$2.00 / $5.00 per 1M tokens · 79 tok/s
- tencentTencent: Hy42026-11-04$0.83 / $2.50 per 1M tokens · 94 tok/s
- OrcaOrcaCyber: MiniMax M3.1 Flash Abliterated / Uncensored2026-10-29$0.80 / $1.60 per 1M tokens · 118 tok/s
- OrcaOrcaCyber: GLM 5.3 Flash Abliterated / Uncensored2026-10-27$0.80 / $1.60 per 1M tokens · 68 tok/s
- OrcaOrcaCyber: Qwen3.8 Flash Next Abliterated / Uncensored2026-10-26$0.80 / $1.60 per 1M tokens · 94 tok/s
- OrcaOrcaCyber: MiMo V2.6 Flash Abliterated / Uncensored2026-10-26$0.80 / $1.60 per 1M tokens · 44 tok/s
- MoonshotMoonshotAI: Kimi K3.1 Pro2026-10-2447Intelligence
- OrcaOrcaCyber: DeepSeek V4.1 Flash Abliterated / Uncensored2026-10-23$0.80 / $1.60 per 1M tokens · 199 tok/s
- minimaxMiniMax: M3.1 Flash2026-10-1929Intelligence
- OrcaOrca: OrcaCyber Ultra 1.02026-10-17$5.00 / $10.00 per 1M tokens · 114 tok/s
- deepseekDeepSeek: DeepSeek V4.1 Flash 10172026-10-1733Intelligence
- DeepSeekDeepSeek: DeepSeek V4.1 Flash 09102026-10-17$0.15 / $0.60 per 1M tokens · 311 tok/s
- DeepSeekDeepSeek: DeepSeek V4.1 Flash2026-10-1740Intelligence
Qwen4-Max costs $4.50 per million output tokens. Gemini 4 Hydrogen, the Google model priced against it, costs $7.50. On Artificial Analysis's own measurements a finished Intelligence Index task runs $3.27 on Qwen4-Max and $3.16 on Hydrogen — a 40% rate advantage on the expensive side of the meter that evaporates into a 3% deficit once you count the tokens each model actually spends. That is the centre of this comparison, and it is invisible on any rate card.
There is a complication before any of the numbers mean anything: Gemini 4 is not a model. It is a line of three separately released checkpoints, and only one of them is priced against Qwen4-Max at all. Getting the row wrong is the most common way this comparison goes bad.
Three Gemini 4s, and which one the question is really about
Google has shipped the Gemini 4 generation three times since 30 September 2026, under element names rather than version numbers. Artificial Analysis scores all three, and they are not close to each other:
• Gemini 4 Argon — announced 30 September 2026, generally available on the paid API from 13 October. 50 on the Index v5.0.1, $2.00 input and $10.00 output, $4.97 per index task, 29.2 minutes per task, 66.4 output tokens per second.
• Gemini 4 Carbon — announced 28 October 2026, distributed through restricted-access programmes rather than open sign-up. 60 on the Index, $2.00 and $10.00, $5.48 per task, 34.2 minutes, 59.9 tokens per second. The highest score in the line and the hardest to get.
• Gemini 4 Hydrogen — shipped 12 November 2026 at $1.50 input and $7.50 output with no cutover date on its card. 54 on the Index, $3.16 per task, 14.9 minutes per task, 172.4 tokens per second — the fastest of the three by a factor of two.
Qwen4-Max, generally available since 16 November 2026, is 43 on the same revision at $1.50 and $4.50 in Singapore, $1.413 and $4.24 in the Global-scope regions. So the price comparison people make — Qwen4-Max against "Gemini 4" — is really Qwen4-Max against Argon, because that is the name the generation launched under and the one most coverage anchors to. The capability comparison, though, points somewhere else entirely, because Hydrogen arrived at exactly Qwen4-Max's input price and beats it by eleven index points.

The one row where they are genuinely level
Hydrogen's input price is $1.50 per million tokens. Qwen4-Max's Singapore input price is $1.50 per million tokens. That is not a rounding coincidence — it is two vendors deciding the same number is the right entry point for a frontier-adjacent model — and it makes every other row decide the purchase.
• Output price — Qwen4-Max $4.50 vs Hydrogen $7.50. A 40% saving on the expensive side of the meter, and the only place Qwen4-Max is straightforwardly cheaper.
• Index, v5.0.1 — Qwen4-Max 43 vs Hydrogen 54, 50 for Argon and 60 for Carbon. Eleven points is a tier change, not a preference.
• Cost per index task — Qwen4-Max $3.27 vs Hydrogen $3.16, Argon $4.97, Carbon $5.48. Qwen4-Max is not the cheap option once tokens-per-answer are counted, despite the friendliest output rate in the group.
• Output tokens to finish the suite — Qwen4-Max 127M, Hydrogen 304M, against a 65.1M median. Both are extremely verbose; Hydrogen is more so, and still finishes cheaper per task because its output is billed at a lower blended rate against fewer reasoning tokens than Carbon.
• Time per index task — Qwen4-Max 32.1 minutes vs Hydrogen 14.9, Argon 29.2, Carbon 34.2. This is the largest functional gap in the table and it points at Google.
• Output speed — 54.5 tokens per second for Qwen4-Max against 172.4 for Hydrogen and 66.4 for Argon.
• Context and output ceiling — 1M tokens for all four, but Qwen4-Max caps non-thinking output at 131,072 and puts thinking plus answer inside a shared 262,144-token budget, while the Gemini 4 pages document a 64K standard output ceiling with long-decode generation extending up to 1M on the Argon and Hydrogen cards.
• Modality — Qwen4-Max takes text, image and video. Argon takes text and image. Hydrogen and Carbon each add speech and video on top of that, which means the one Gemini 4 that matches Qwen4-Max on price is also the one that already covers its modality gap.
• Weights and access — closed on all four. Carbon is the only one that needs an application.

What a single request costs, before the harness gets involved
Take a 150,000-token prompt with a 4,000-token answer, priced at list, thinking off, Singapore for Qwen4-Max and standard rates for Hydrogen.
On Qwen4-Max the input is $0.225 and the output $0.018, for $0.243. On Hydrogen the input is $0.225 and the output $0.030, for $0.255. Qwen4-Max wins by five cents, or 5%.
Now let the model think. Artificial Analysis measures 104,940 output tokens per index task for Qwen4-Max, of which 68,914 are reasoning, against a much shorter answer. Re-run the same 150,000-token prompt with a 100,000-token thinking budget: Qwen4-Max bills $0.225 plus $0.45, or $0.675. Hydrogen at the same thinking volume bills $0.225 plus $0.75, or $0.975. Qwen4-Max wins by 31%.
So the arithmetic on a single request still favours Qwen4-Max, and the per-task figure on the tracker favours Hydrogen. Both are correct, and they do not reconcile by multiplying output tokens by the output rate — Hydrogen generated 304M tokens to Qwen4-Max's 127M, 2.4× as many, at 1.67× the price, which by that arithmetic should make it roughly four times dearer per run rather than slightly cheaper. The published per-task figure is a weighted blend of input, cached input, reasoning and answer tokens across nine evaluations, and the gap between the naive multiplication and the measured result is the part to be suspicious of. Until someone publishes the token split for both runs, treat the two per-task numbers as a measured comparison you cannot reconstruct from the rate cards, and treat any article that derives one from the other as arithmetic dressed up as evidence.
The clock nobody has priced yet
Argon's card is introductory. On 1 December 2026 it moves from $2.00 / $10.00 to $4.00 / $20.00, with caching rates doubling alongside. That is thirteen days from this page, and it changes the shape of the comparison more than any benchmark delta in it.
Today, Qwen4-Max against Argon is a 25% input and 55% output saving for a seven-point capability deficit. From 1 December it becomes 62.5% and 77.5% for the same deficit. A team that benchmarks in late November and signs in December is evaluating a different product from the one it gets.
Hydrogen, by contrast, carries no cutover date on its card, and Carbon's restricted distribution means its $2.00 / $10.00 is not a public decision at all. So the honest reading of "Qwen4-Max vs Gemini 4" depends on which of three clocks you are standing in front of: the Argon clock, which is running; the Hydrogen card, which is not; and the Qwen4-Max card, which is also not — Alibaba has published one flat tier from zero to the context limit with no long-context surcharge and no promotional end date.
What Google's models do that Alibaba's does not, and the reverse
The Gemini 4 line's practical advantage is not the score. It is that all three checkpoints have been running long enough to have published throughput, and 172 tokens per second on Hydrogen is a different class of product from 54.5. For anything in front of a user — an IDE assistant, a support agent, a document tool with a spinner — that difference is the product.
Qwen4-Max's advantages are narrower and more concrete. A single flat price tier across the whole 1M window is one: Google's long-decode generation above the standard output ceiling is documented as an extension rather than a flat rate. Protocol breadth is another — Chat Completions, Responses, Anthropic Messages and DashScope against a single vendor wire format, which is the difference between changing a base URL and changing a client. Modality is not, and anyone who tells you video input is a Qwen-only row has not read the Hydrogen card, which lists speech and video alongside text and image.
There is also the regional question. Qwen4-Max is metered in six regions including Frankfurt and US Virginia, at $1.413 / $4.24 outside Singapore, with batch inference available in Beijing only. Google's Gemini 4 pages publish a single global card. For a team with EU data-residency requirements the Alibaba figure to quote is the Frankfurt one, not the Singapore headline, and it is 5.8% lower.
Calling all four from one place
OrcaRouter carries two of the four: Gemini 4 Argon and Gemini 4 Carbon are in the catalogue as google/gemini-4-argon and google/gemini-4-carbon at Google's list prices with 0% markup passed through, so the 1 December Argon increase lands here the same day it lands on Google's card rather than at the next billing cycle. Qwen4-Max is not hosted here — it is available through Alibaba's own Model Studio API and several third-party platforms.

The useful pattern for this specific comparison is the one the pricing structure suggests: Carbon for the hard cases, Argon as the general default until its price moves, and a Qwen-class model for the video and long-cached-prefix work where the Gemini line has no answer. That is three model choices on one request path, which is what the routing DSL exists for — classify the request, send it to the model the row says it belongs on, and let automatic failover cover the day a provider has a bad hour. A single key and one OpenAI-compatible endpoint make the A/B test cheap enough to run before the December cutover instead of after it.
Read as a family, Gemini 4 is not one opponent for Qwen4-Max but three, and Qwen4-Max loses the score row to all of them while winning the output-price row against all of them. The number that actually decides it is $3.27 against $3.16 — a per-task tie that a rate-card comparison would have called a 40% win — followed by 32.1 minutes against 14.9, which no amount of token pricing fixes. The date to circle is 1 December, when Argon stops being the cheap Google option and the only honest comparison left is against Hydrogen, at which point the eleven-point gap and the 6.5× speed gap are the entire argument.
