Compare / Gemini 3.1 Pro Preview vs MiniMax M3.1 Flash
Gemini 3.1 Pro Preview vs MiniMax M3.1 Flash
Which is cheaper, which is more capable, and which does better at each kind of work, from 82 results on 5 sources that measured both models.
GoogleGemini 3.1 Pro Preview
MiniMaxMiniMax M3.1 Flash
Gemini 3.1 Pro Preview wins 4410 too close to callMiniMax M3.1 Flash wins 28
Of the 82 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 20 October 2026 (catalogue release 2026-10-20-96eb40c3ec63). See also Google vs MiniMax, every model from both labs.
Gemini 3.1 Pro Preview or MiniMax M3.1 Flash? The short answer
They score about the same, so pick MiniMax M3.1 Flash: it costs 17 times less per token.
- Spring Prompt overall: Gemini 3.1 Pro Preview 48, MiniMax M3.1 Flash 47 (out of 100; how the score works).
- MiniMax M3.1 Flash is cheaper: $0.26 against $4.50 per million tokens, 17 times less.
- MiniMax M3.1 Flash scores higher on the Artificial Analysis Intelligence Index: 42.4 against 29.7.
- MiniMax M3.1 Flash is newer: released 19 Oct 2026, against 19 Feb 2026 for Gemini 3.1 Pro Preview.
- On our own benchmarks, on the results both models have: Gemini 3.1 Pro Preview does better for user surveys and marketing planning; MiniMax M3.1 Flash for product listings and decks from an analysis.
- On third-party benchmarks, on the results both models have: MiniMax M3.1 Flash does better for agents and tool use and reasoning and knowledge.
Head to head
48
Spring Prompt overall
47
29.7
Intelligence Index
42.4
$4.50
Price per million tokens, blendedlower is better
$0.26
130
Output speed, tokens a second
159
1,048,576
Context window, tokens
1,048,576
Every fact side by side
| Gemini 3.1 Pro Preview | MiniMax M3.1 Flash | |
|---|---|---|
| Price per million tokens | $2.00 in · $12.00 out | $0.15 in · $0.60 out |
| Blended (3 in : 1 out) | $4.50 | $0.26 |
| Cheapest host, blended | $4.50 (Google) | – |
| Cached input, per million | $0.20 cached input | $0.015 cached input |
| Intelligence Index | 29.7 | 42.4 |
| Spring Prompt overall | 48 | 47 |
| Output speed, tokens a second | 130 | 159 |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Released | 19 Feb 2026 | 19 Oct 2026 (newer) |
| Weights | Closed | Closed |
| Developer based in | the US | China |
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
11 of 24
7 of 24
Decks from an analysisTurning a finished analysis into a deck you could present as it is
19 of 24
14 of 24
User surveysPlanning a user survey and reading its results without being misled
8 of 13
13 of 13
Marketing planningPlanning a year of ad spend without overspending
10 of 24
12 of 24
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
64 of 168
32 of 168
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
36 of 51
not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge
27 of 202
18 of 202
WritingWhat people prefer in blind comparisons, and judged writing quality
9 of 189
not measured
Sticking to the factsSummarising without inventing things, and factual answers
35 of 153
not measured
Speed under pressureGood decisions against a real clock (fast chess)
5 of 27
not measured
From our research
- Gemini 4 Argon review: Opus-level product data at a quarter of the cost, until you ask it to sell 14 Oct 2026
- BulletBench: we made 23 AI models play speed chess, and the clock was the judge 3 Jul 2026
- Claude vs Gemini vs GPT in a 12-Month Marketing Simulation 25 Mar 2026
- The AI Marketing Benchmark That Punishes Plausible-Sounding Strategy 20 Mar 2026
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.
| Measure | Gemini 3.1 Pro Preview | MiniMax M3.1 Flash |
|---|---|---|
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 19 · 1 of 24 measured | $0.0355 | $0 |
| Reliably publish-ready% of products, higher is better 11 · 9 of 24 measured | 44.6% | 50.0% |
| Reliably publish-ready · sales brief% of products, higher is better 11 · 6 of 24 measured | 16.1% | 37.5% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 14 · 1 of 24 measured | $0.14 | $0 |
| 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. 19 · 14 of 24 measured | 788 | 1,090 |
| Accurate decks% of tasks, higher is better 9 · 9 of 24 measured | 50.0% | 50.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 10 · 12 of 24 measured | 43.5 | 40.3 |
| Audience scorescore out of 100, higher is better 6 · 12 of 24 measured | 67.5 | 60.8 |
| Business scorescore out of 100, higher is better 12 · 10 of 24 measured | 35.9 | 38.9 |
| 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. 8 · 13 of 13 measured | 797 | 441 |
| Analysis scorepoints of 100, higher is better 7 · 13 of 13 measured | 66.4 | 7.90 |
| 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. 9 · 13 of 13 measured | 34.0% | 15.9% |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 57 · 21 of 249 measured | 29.7 | 42.4 |
| Humanity's Last Exam% of questions, higher is better 18 · 14 of 248 measured | 47.0% | 48.0% |
| Terminal-Bench 4.0% of tasks, higher is better 49 · 22 of 118 measured | 4.0% | 25.8% |
Show all 82 results (67 more)
| Measure | Gemini 3.1 Pro Preview | MiniMax M3.1 Flash |
|---|---|---|
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 17 · 1 of 24 measured | 99.4% | 100.0% |
| Channel rules broken% of products, lower is better 12 · 12 of 24 measured | 0.6% | 0.6% |
| Channel rules broken · sales brief% of products, lower is better 1 · 1 of 24 measured | 0.0% | 0.0% |
| Claims to check% of products, lower is better 19 · 1 of 24 measured | 1.2% | 0.0% |
| Claims to check · sales brief% of products, lower is better 21 · 1 of 24 measured | 10.1% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 24 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 12 · 17 of 24 measured | 96.9% | 95.8% |
| Content quality · sales brief% of checks, higher is better 12 · 17 of 24 measured | 97.2% | 96.0% |
| Decision accuracy% of decisions, higher is better 11 · 16 of 24 measured | 96.6% | 95.6% |
| Failed outputs% of products, lower is better 1 · 22 of 24 measured | 0.0% | 1.8% |
| Failed outputs · sales brief% of products, lower is better 1 · 22 of 24 measured | 0.0% | 1.8% |
| Field accuracy% of missing fields, higher is better 2 · 19 of 24 measured | 95.4% | 92.0% |
| Invented values% of filled values, lower is better 12 · 22 of 24 measured | 1.5% | 3.8% |
| Missing UK information% of products, lower is better 1 · 1 of 24 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 24 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 17 · 16 of 24 measured | 7.7% | 7.1% |
| Not findable · sales brief% of products, lower is better 17 · 12 of 24 measured | 6.0% | 4.2% |
| Publish-ready listings% of products, higher is better 13 · 11 of 24 measured | 58.9% | 63.1% |
| Publish-ready listings · sales brief% of products, higher is better 12 · 7 of 24 measured | 30.4% | 57.1% |
| Time per productseconds, lower is better 18 · 19 of 24 measured | 22.2 | 32.0 |
| Unsupported claimsclaims per product, lower is better 12 · 6 of 24 measured | 0.66 | 0.39 |
| Unsupported claims% of products, lower is better 13 · 5 of 24 measured | 26.8% | 7.7% |
| Unsupported claims · sales briefclaims per product, lower is better 14 · 6 of 24 measured | 2.13 | 0.60 |
| Unsupported claims · sales brief% of products, lower is better 14 · 5 of 24 measured | 67.3% | 16.1% |
| Wrong attributes% of products, lower is better 5 · 20 of 24 measured | 14.9% | 22.6% |
| Wrong attributes · sales brief% of products, lower is better 10 · 20 of 24 measured | 16.1% | 21.4% |
| Wrong category or variant% of products, lower is better 5 · 20 of 24 measured | 1.8% | 4.2% |
| Wrong category or variant · sales brief% of products, lower is better 5 · 21 of 24 measured | 1.8% | 4.8% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 12 · 24 of 24 measured | 16.7% | 33.3% |
| Clean layout% of tasks, higher is better 12 · 23 of 24 measured | 50.0% | 0.0% |
| Design quality% of the maximum, higher is better 19 · 21 of 24 measured | 51.4% | 46.2% |
| Draft figure quoted% of tasks, lower is better 1 · 12 of 24 measured | 0.0% | 33.3% |
| Findings missing% of tasks, lower is better 1 · 24 of 24 measured | 0.0% | 50.0% |
| 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. 19 · 15 of 24 measured | 34.0% | 48.6% |
| Layout defects% of tasks, lower is better 12 · 23 of 24 measured | 50.0% | 100.0% |
| Misleading metric used% of tasks, lower is better 1 · 24 of 24 measured | 0.0% | 33.3% |
| Presentable decks% of tasks, higher is better 10 · 12 of 24 measured | 16.7% | 0.0% |
| Recommendation late or wrong% of tasks, lower is better 17 · 13 of 24 measured | 50.0% | 33.3% |
| Slides needing work% of tasks, lower is better 21 · 21 of 24 measured | 83.3% | 83.3% |
| Time per deckseconds, lower is better 8 · 20 of 24 measured | 78.0 | 330 |
| Unsupported claims% of tasks, lower is better 1 · 21 of 24 measured | 0.0% | 50.0% |
| Unsupported numbers% of tasks, lower is better 1 · 24 of 24 measured | 0.0% | 50.0% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 6 · 17 of 24 measured | 33.9 | 17.3 |
| Contribution profitsimulated US dollars, higher is better 13 · 5 of 24 measured | $507,736 | $693,379 |
| Cost of a runUS dollars, lower is better 18 · 1 of 24 measured | $0.52 | $0 |
| Months over budgetmonths of 12, lower is better 1 · 1 of 24 measured | 0 | 0 |
| Planning scorescore out of 100, higher is better 8 · 21 of 24 measured | 54.9 | 54.2 |
| Return on ad spendprofit per $1 spent, higher is better 13 · 9 of 24 measured | $0.42 | $0.52 |
| SurveyBench measured by Spring Prompt · 15 results | ||
| Leading or double-barrelled questions% of tasks, lower is better 6 · 12 of 13 measured | 44.4% | 77.8% |
| Mistaken next steps% of tasks, lower is better 7 · 13 of 13 measured | 33.3% | 100.0% |
| Numbers right% of answers, higher is better 1 · 1 of 13 measured | 100.0% | 100.0% |
| Objectives unanswerable% of tasks, lower is better 1 · 13 of 13 measured | 0.0% | 66.7% |
| Precision ignored% of tasks, lower is better 8 · 10 of 13 measured | 55.6% | 66.7% |
| Research scorepoints of 100, higher is better 7 · 13 of 13 measured | 71.3 | 16.4 |
| Sound research% of tasks, higher is better 6 · 6 of 13 measured | 0.0% | 0.0% |
| Stakeholder's view presumed% of tasks, lower is better 7 · 12 of 13 measured | 22.2% | 55.6% |
| Survey plan scorepoints of 100, higher is better 10 · 13 of 13 measured | 82.9 | 36.1 |
| Survey structure% of tasks, lower is better 7 · 13 of 13 measured | 22.2% | 77.8% |
| Traps handled% of traps, higher is better 9 · 8 of 13 measured | 71.0% | 75.0% |
| Traps missed% of tasks, lower is better 7 · 10 of 13 measured | 88.9% | 100.0% |
| Unsupported findings% of tasks, lower is better 7 · 10 of 13 measured | 77.8% | 100.0% |
| Wrong numbers% of tasks, lower is better 1 · 10 of 13 measured | 0.0% | 88.9% |
| Wrong recommendation% of tasks, lower is better 8 · 13 of 13 measured | 44.4% | 100.0% |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| Long-context reasoning (AA-LCR)% of questions, higher is better 25 · 45 of 239 measured | 82.0% | 79.7% |
| Output speedtokens per second, higher is better 36 · 21 of 78 measured | 130 | 159 |
| SciCode% of problems, higher is better 13 · 36 of 119 measured | 58.7% | 53.4% |
| Time to first answer tokenseconds, lower is better 57 · 51 of 78 measured | 14.7 | 13.3 |
Compare them with other models
- Google vs MiniMaxEvery model from both labs, best against best
- Gemini 3.1 Pro Preview vs Claude Opus 5.5Most intelligent at $3–10 per million tokens
- Gemini 3.1 Pro Preview vs Inkling SmallFastest with similar intelligence
- MiniMax M3.1 Flash vs Claude Haiku 5.5Cheapest with similar intelligence
- MiniMax M3.1 Flash vs DeepSeek-V4.1-FlashFastest with similar intelligence
- MiniMax M3.1 Flash vs Claude Opus 5.5Most intelligent available today
Quick answers
Is Gemini 3.1 Pro Preview better than MiniMax M3.1 Flash?
They score about the same, so pick MiniMax M3.1 Flash: it costs 17 times less per token.
Which is cheaper, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
MiniMax M3.1 Flash is cheaper: $0.26 against $4.50 per million tokens, 17 times less.
Which is faster, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
MiniMax M3.1 Flash writes about 159 tokens a second on its usual API, against 130 for Gemini 3.1 Pro Preview (Artificial Analysis).
Which is better for product listings, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
MiniMax M3.1 Flash ranks 7 of 24 models on sale for turning a sparse product feed and photos into listings that can go live, against 11 for Gemini 3.1 Pro Preview.
Which is better for decks from an analysis, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
MiniMax M3.1 Flash ranks 14 of 24 models on sale for turning a finished analysis into a deck you could present as it is, against 19 for Gemini 3.1 Pro Preview.
Which is better for user surveys, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
Gemini 3.1 Pro Preview ranks 8 of 13 models on sale for planning a user survey and reading its results without being misled, against 13 for MiniMax M3.1 Flash.
Which is better for marketing planning, Gemini 3.1 Pro Preview or MiniMax M3.1 Flash?
Gemini 3.1 Pro Preview ranks 10 of 24 models on sale for planning a year of ad spend without overspending, against 12 for MiniMax M3.1 Flash.
Run Gemini 3.1 Pro Preview and MiniMax M3.1 Flash on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.