Compare / Qwen4-Max vs Gemini 3.8 Flash

Qwen4-Max vs Gemini 3.8 Flash

Which is cheaper, which is more capable, and which does better at each kind of work, from 88 results on 5 sources that measured both models.

AlibabaQwen4-Max $1.50 in · $4.50 out
GoogleGemini 3.8 Flash Released 2 Sep 2026 · $1.50 in · $7.50 out
Qwen4-Max wins 4118 too close to callGemini 3.8 Flash wins 29

Of the 88 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Alibaba vs Google, every model from both labs.

Qwen4-Max or Gemini 3.8 Flash? The short answer

They score about the same, so pick Qwen4-Max: it costs 33% less per product listing on CatalogBench.

  • Spring Prompt overall: Qwen4-Max 56, Gemini 3.8 Flash 58 (out of 100; how the score works).
  • Qwen4-Max is cheaper to run: $0.007 against $0.010 per product listing on CatalogBench (33% less).
  • On our own benchmarks, on the results both models have: Qwen4-Max does better for product listings, decks from an analysis and user surveys; Gemini 3.8 Flash for marketing planning and speed under pressure.

Head to head

56
Spring Prompt overall
58
$2.25
Price per million tokens, blendedlower is better
$3.00
1,000,000
Context window, tokens
1,048,576

Every fact side by side

Qwen4-MaxGemini 3.8 Flash
Price per million tokens$1.50 in · $4.50 out$1.50 in · $7.50 out
Blended (3 in : 1 out)$2.25$3.00
Cheapest host, blended–$1.50 (Google)
Cost per product listing (CatalogBench)$0.007$0.010
Intelligence Index–40.9
Spring Prompt overall5658
Output speed, tokens a second–141
Context window1,000,000 tokens1,048,576 tokens
Released–2 Sep 2026
WeightsClosedClosed
Developer based inChinathe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live 19 of 32 22 of 32
Decks from an analysisTurning a finished analysis into a deck you could present as it is 1 of 32 26 of 32
User surveysPlanning a user survey and reading its results without being misled 5 of 23 16 of 23
Marketing planningPlanning a year of ad spend without overspending 15 of 32 11 of 32
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls not measured 35 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 11 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge not measured 15 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality not measured 3 of 169
Sticking to the factsSummarising without inventing things, and factual answers not measured 12 of 139
Speed under pressureGood decisions against a real clock (fast chess) 26 of 35 13 of 35

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureQwen4-MaxGemini 3.8 Flash
BulletBench measured by Spring Prompt · 3 results
Cost per game · Blitz 3+2US dollars, lower is better 7 · 11 of 20 measured $0.0400 $0.0795
Cost per game · Bullet 60sUS dollars, lower is better 42 · 43 of 59 measured $0.0150low reasoning $0.0154
Ladder Elo · Bullet 60sladder Elo, higher is better 51 · 1 of 59 measured 147low reasoning 1,068low reasoning
CatalogBench measured by Spring Prompt · 3 results
Cost per productUS dollars, lower is better 13 · 18 of 33 measured $0.0069 $0.0103
Reliably publish-ready% of products, higher is better 19 · 20 of 32 measured 39.3% 35.7%
Reliably publish-ready · sales brief% of products, higher is better 21 · 23 of 32 measured 8.9% 3.6%
DeckBench measured by Spring Prompt · 3 results
Cost per deckUS dollars, lower is better 23 · 15 of 33 measured $0.18 $0.0613
Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 1 · 26 of 33 measured 1,585 788
Accurate decks% of tasks, higher is better 19 · 24 of 33 measured 33.3% 16.7%
ROASBench measured by Spring Prompt · 3 results
Overall scorescore out of 100, higher is better 15 · 11 of 33 measured 45.8 51.4
Audience scorescore out of 100, higher is better 23 · 10 of 33 measured 55.1 67.1
Business scorescore out of 100, higher is better 13 · 11 of 33 measured 48.5 51.0
SurveyBench measured by Spring Prompt · 3 results
Analysis ratingrating, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 5 · 16 of 24 measured 1,395 875
Cost per taskUS dollars, lower is better 17 · 14 of 24 measured $0.20 $0.0943
Analysis scorepoints of 100, higher is better 8 · 15 of 24 measured 88.9 74.7
Show all 88 results (73 more)
MeasureQwen4-MaxGemini 3.8 Flash
BulletBench measured by Spring Prompt · 7 results
Games lost on time · Blitz 3+2% of games, lower is better 20 · 7 of 20 measured 87.5% 12.5%
Games lost on time · Bullet 60s% of games, lower is better 54 · 39 of 59 measured 87.5%low reasoning 50.0%low reasoning
Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 20 measured 0.0% 0.0%
Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 59 measured 0.0%low reasoning 0.0%
Ladder Elo · Blitz 3+2ladder Elo, higher is better 20 · 2 of 20 measured 147 1,126
Median move time · Blitz 3+2milliseconds, lower is better 20 · 9 of 20 measured 8.6 s 4.0 s
Median move time · Bullet 60smilliseconds, lower is better 58 · 46 of 59 measured 4.4 slow reasoning 1.8 slow reasoning
CatalogBench measured by Spring Prompt · 28 results
Channel compliance% of products, higher is better 1 · 25 of 33 measured 100.0% 99.4%
Channel rules broken% of products, lower is better 1 · 20 of 33 measured 0.0% 0.6%
Channel rules broken · sales brief% of products, lower is better 25 · 17 of 33 measured 1.2% 0.6%
Claims to check% of products, lower is better 27 · 1 of 31 measured 1.8% 0.0%
Claims to check · sales brief% of products, lower is better 23 · 26 of 31 measured 3.0% 5.4%
Conflicts caught% of conflicts, higher is better 1 · 1 of 33 measured 100.0% 100.0%
Content quality% of checks, higher is better 15 · 9 of 33 measured 97.5% 97.7%
Content quality · sales brief% of checks, higher is better 12 · 14 of 33 measured 97.9% 97.6%
Decision accuracy% of decisions, higher is better 17 · 19 of 33 measured 96.8% 96.4%
Failed outputs% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Failed outputs · sales brief% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Field accuracy% of missing fields, higher is better 21 · 24 of 33 measured 92.9% 92.7%
Invented values% of filled values, lower is better 27 · 4 of 33 measured 2.9% 0.3%
Missing UK information% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Missing UK information · sales brief% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Not findable% of products, lower is better 9 · 27 of 33 measured 3.0% 8.3%
Not findable · sales brief% of products, lower is better 20 · 28 of 33 measured 4.2% 8.9%
Publish-ready listings% of products, higher is better 20 · 21 of 33 measured 56.6% 54.8%
Publish-ready listings · sales brief% of products, higher is better 20 · 23 of 33 measured 25.6% 16.7%
Time per productseconds, lower is better 21 · 19 of 33 measured 20.9 18.7
Unsupported claims% of products, lower is better 20 · 24 of 33 measured 22.0% 31.6%
Unsupported claimsclaims per product, lower is better 20 · 22 of 33 measured 0.70 0.79
Unsupported claims · sales brief% of products, lower is better 22 · 28 of 33 measured 70.2% 81.6%
Unsupported claims · sales briefclaims per product, lower is better 23 · 24 of 33 measured 2.56 3.08
Wrong attributes% of products, lower is better 26 · 11 of 33 measured 20.2% 15.5%
Wrong attributes · sales brief% of products, lower is better 27 · 2 of 33 measured 21.4% 11.9%
Wrong category or variant% of products, lower is better 6 · 27 of 33 measured 1.8% 3.6%
Wrong category or variant · sales brief% of products, lower is better 18 · 23 of 33 measured 2.4% 3.0%
DeckBench measured by Spring Prompt · 14 results
Caveat dropped% of tasks, lower is better 15 · 1 of 33 measured 16.7% 0.0%
Clean layout% of tasks, higher is better 24 · 24 of 33 measured 16.7% 16.7%
Design quality% of the maximum, higher is better 4 · 19 of 33 measured 83.4% 61.7%
Draft figure quoted% of tasks, lower is better 13 · 26 of 33 measured 16.7% 50.0%
Findings missing% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 3 · 26 of 33 measured 86.3% 38.7%
Layout defects% of tasks, lower is better 24 · 24 of 33 measured 83.3% 83.3%
Misleading metric used% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Presentable decks% of tasks, higher is better 15 · 15 of 33 measured 0.0% 0.0%
Recommendation late or wrong% of tasks, lower is better 1 · 20 of 33 measured 0.0% 33.3%
Slides needing work% of tasks, lower is better 10 · 10 of 33 measured 33.3% 33.3%
Time per deckseconds, lower is better 31 · 13 of 33 measured 489 93.0
Unsupported claims% of tasks, lower is better 24 · 30 of 33 measured 33.3% 50.0%
Unsupported numbers% of tasks, lower is better 26 · 26 of 33 measured 16.7% 16.7%
ROASBench measured by Spring Prompt · 6 results
Consistency scorescore out of 100, higher is better 22 · 6 of 33 measured 25.5 36.4
Contribution profitsimulated US dollars, higher is better 7 · 5 of 33 measured $775,520 $842,432
Cost of a runUS dollars, lower is better 26 · 15 of 33 measured $0.69 $0.16
Months over budgetmonths of 12, lower is better 1 · 1 of 33 measured 0 0
Planning scorescore out of 100, higher is better 31 · 21 of 33 measured 54.0 54.7
Return on ad spendprofit per $1 spent, higher is better 7 · 12 of 33 measured $0.68 $0.65
SurveyBench measured by Spring Prompt · 18 results
Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 5 · 6 of 24 measured 69.9% 62.5%
Leading or double-barrelled questions% of tasks, lower is better 3 · 21 of 24 measured 11.1% 66.7%
Mistaken next steps% of tasks, lower is better 5 · 5 of 24 measured 11.1% 11.1%
Numbers right% of answers, higher is better 16 · 1 of 24 measured 99.0% 100.0%
Objectives unanswerable% of tasks, lower is better 1 · 1 of 24 measured 0.0% 0.0%
Output tokens per tasktokens, lower is better 8 · 3 of 14 measured 39,999 21,981
Precision ignored% of tasks, lower is better 1 · 18 of 24 measured 0.0% 44.4%
Research scorepoints of 100, higher is better 8 · 15 of 24 measured 90.8 77.6
Sound research% of tasks, higher is better 4 · 15 of 24 measured 33.3% 0.0%
Stakeholder's view presumed% of tasks, lower is better 1 · 15 of 24 measured 0.0% 22.2%
Survey plan scorepoints of 100, higher is better 5 · 18 of 24 measured 95.2 84.4
Survey structure% of tasks, lower is better 11 · 11 of 24 measured 22.2% 22.2%
Time per taskseconds, lower is better 23 · 8 of 24 measured 690 131
Traps handled% of traps, higher is better 2 · 18 of 24 measured 98.6% 78.3%
Traps missed% of tasks, lower is better 2 · 16 of 24 measured 11.1% 88.9%
Unsupported findings% of tasks, lower is better 11 · 14 of 24 measured 55.6% 77.8%
Wrong numbers% of tasks, lower is better 13 · 1 of 24 measured 11.1% 0.0%
Wrong recommendation% of tasks, lower is better 12 · 1 of 24 measured 22.2% 0.0%

Quick answers

Is Qwen4-Max better than Gemini 3.8 Flash?

They score about the same, so pick Qwen4-Max: it costs 33% less per product listing on CatalogBench.

Which is cheaper, Qwen4-Max or Gemini 3.8 Flash?

Qwen4-Max is cheaper to run: $0.007 against $0.010 per product listing on CatalogBench (33% less).

Which is better for product listings, Qwen4-Max or Gemini 3.8 Flash?

Qwen4-Max ranks 19 of 32 models on sale for turning a sparse product feed and photos into listings that can go live, against 22 for Gemini 3.8 Flash.

Which is better for decks from an analysis, Qwen4-Max or Gemini 3.8 Flash?

Qwen4-Max ranks 1 of 32 models on sale for turning a finished analysis into a deck you could present as it is, against 26 for Gemini 3.8 Flash.

Which is better for user surveys, Qwen4-Max or Gemini 3.8 Flash?

Qwen4-Max ranks 5 of 23 models on sale for planning a user survey and reading its results without being misled, against 16 for Gemini 3.8 Flash.

Which is better for marketing planning, Qwen4-Max or Gemini 3.8 Flash?

Gemini 3.8 Flash ranks 11 of 32 models on sale for planning a year of ad spend without overspending, against 15 for Qwen4-Max.

Which is better for speed under pressure, Qwen4-Max or Gemini 3.8 Flash?

Gemini 3.8 Flash ranks 13 of 35 models on sale for good decisions against a real clock (fast chess), against 26 for Qwen4-Max.

Which has the bigger context window, Qwen4-Max or Gemini 3.8 Flash?

Gemini 3.8 Flash: 1,048,576 tokens, against 1,000,000 for Qwen4-Max.

Run Qwen4-Max and Gemini 3.8 Flash on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.