Compare / Mistral Medium 3.5 vs Inkling 2 Mini

Mistral Medium 3.5 vs Inkling 2 Mini

Which is cheaper, which is more capable, and which does better at each kind of work, from 77 results on 4 sources that measured both models.

Mistral AIMistral Medium 3.5 Released 28 Apr 2026 · $1.50 in · $7.50 out
Thinking MachinesInkling 2 Mini Released 20 Nov 2026 · $0.091 in · $0.36 out
Mistral Medium 3.5 wins 259 too close to callInkling 2 Mini wins 43

Of the 77 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 21 November 2026 (catalogue release 2026-11-21-af3a2faf6a5c).

Mistral Medium 3.5 or Inkling 2 Mini? The short answer

Pick Inkling 2 Mini: it scores higher (Spring Prompt overall 21 against 15) and costs less.

  • Spring Prompt overall: Mistral Medium 3.5 15, Inkling 2 Mini 21 (out of 100; how the score works).
  • Inkling 2 Mini is cheaper to run: $0.001 against $0.008 per product listing on CatalogBench (5.4 times less).
  • Inkling 2 Mini is newer: released 20 Nov 2026, against 28 Apr 2026 for Mistral Medium 3.5.
  • On our own benchmarks, on the results both models have: Inkling 2 Mini does better for decks from an analysis and marketing planning.

Head to head

15
Spring Prompt overall
21
$3.00
Price per million tokens, blendedlower is better
$0.16
262,144
Context window, tokens
1,048,576

Every fact side by side

Mistral Medium 3.5Inkling 2 Mini
Price per million tokens$1.50 in · $7.50 out$0.091 in · $0.36 out
Blended (3 in : 1 out)$3.00$0.16
Cheapest host, blended$3.00 (Mistral)–
Cost per product listing (CatalogBench)$0.008$0.001
Cached input, per million$0.15 cached input–
Intelligence Index14.2–
Spring Prompt overall1521
Output speed, tokens a second157–
Context window262,144 tokens1,048,576 tokens
Released28 Apr 202620 Nov 2026 (newer)
WeightsClosedOpen
Developer based inFrancethe 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 26 of 36 28 of 36
Decks from an analysisTurning a finished analysis into a deck you could present as it is 35 of 36 29 of 36
User surveysPlanning a user survey and reading its results without being misled 24 of 27 26 of 27
Marketing planningPlanning a year of ad spend without overspending 36 of 36 29 of 36
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 100 of 159 not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge 123 of 187 not measured
WritingWhat people prefer in blind comparisons, and judged writing quality 85 of 169 not measured
Sticking to the factsSummarising without inventing things, and factual answers 98 of 139 not measured
Speed under pressureGood decisions against a real clock (fast chess) 13 of 38 39 of 39

Their slides, side by side

The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.

First slide of Mistral Medium 3.5's deck for Hearthside Coffee
Mistral Medium 3.5 · deck rating 588 · needs work
First slide of Inkling 2 Mini's deck for Hearthside Coffee
Inkling 2 Mini · deck rating 792 · needs work

Every slide, side by side →

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.

MeasureMistral Medium 3.5Inkling 2 Mini
CatalogBench measured by Spring Prompt · 3 results
Cost per productUS dollars, lower is better 16 · 5 of 37 measured $0.0080 $0.0015
Reliably publish-ready% of products, higher is better 30 · 28 of 36 measured 17.9% 23.2%
Reliably publish-ready · sales brief% of products, higher is better 20 · 25 of 36 measured 10.7% 5.4%
DeckBench measured by Spring Prompt · 3 results
Cost per deckUS dollars, lower is better 12 · 9 of 37 measured $0.0448 $0.0223
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. 36 · 29 of 37 measured 588 792
Accurate decks% of tasks, higher is better 31 · 31 of 37 measured 0.0% 0.0%
ROASBench measured by Spring Prompt · 3 results
Overall scorescore out of 100, higher is better 37 · 29 of 37 measured 15.2 29.9
Audience scorescore out of 100, higher is better 37 · 28 of 37 measured 28.5 54.0
Business scorescore out of 100, higher is better 37 · 29 of 37 measured 3.4 24.1
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. 25 · 27 of 28 measured 592 567
Cost per taskUS dollars, lower is better 13 · 4 of 28 measured $0.0730 $0.0168
Analysis scorepoints of 100, higher is better 21 · 25 of 28 measured 56.7 16.5
Show all 77 results (65 more)
MeasureMistral Medium 3.5Inkling 2 Mini
CatalogBench measured by Spring Prompt · 28 results
Channel compliance% of products, higher is better 1 · 34 of 37 measured 100.0% 98.8%
Channel rules broken% of products, lower is better 1 · 35 of 37 measured 0.0% 3.0%
Channel rules broken · sales brief% of products, lower is better 28 · 35 of 37 measured 1.2% 3.6%
Claims to check% of products, lower is better 33 · 1 of 35 measured 3.6% 0.0%
Claims to check · sales brief% of products, lower is better 33 · 26 of 35 measured 25.6% 1.8%
Conflicts caught% of conflicts, higher is better 36 · 1 of 37 measured 90.9% 100.0%
Content quality% of checks, higher is better 36 · 32 of 37 measured 94.1% 95.3%
Content quality · sales brief% of checks, higher is better 34 · 33 of 37 measured 94.1% 94.1%
Decision accuracy% of decisions, higher is better 35 · 29 of 37 measured 92.6% 95.3%
Failed outputs% of products, lower is better 1 · 1 of 37 measured 0.0% 0.0%
Failed outputs · sales brief% of products, lower is better 1 · 31 of 37 measured 0.0% 1.2%
Field accuracy% of missing fields, higher is better 33 · 18 of 37 measured 91.7% 93.7%
Invented values% of filled values, lower is better 1 · 33 of 37 measured 0.0% 3.8%
Missing UK information% of products, lower is better 1 · 1 of 37 measured 0.0% 0.0%
Missing UK information · sales brief% of products, lower is better 1 · 1 of 37 measured 0.0% 0.0%
Not findable% of products, lower is better 36 · 18 of 37 measured 14.3% 4.2%
Not findable · sales brief% of products, lower is better 33 · 17 of 37 measured 10.1% 3.6%
Publish-ready listings% of products, higher is better 33 · 26 of 37 measured 29.8% 47.0%
Publish-ready listings · sales brief% of products, higher is better 26 · 22 of 37 measured 17.3% 27.4%
Time per productseconds, lower is better 2 · 30 of 37 measured 4.90 40.7
Unsupported claimsclaims per product, lower is better 35 · 33 of 37 measured 1.74 1.24
Unsupported claims% of products, lower is better 35 · 26 of 37 measured 49.4% 27.4%
Unsupported claims · sales brief% of products, lower is better 28 · 22 of 37 measured 76.8% 61.9%
Unsupported claims · sales briefclaims per product, lower is better 30 · 22 of 37 measured 3.55 2.01
Wrong attributes% of products, lower is better 35 · 33 of 37 measured 31.0% 24.4%
Wrong attributes · sales brief% of products, lower is better 36 · 30 of 37 measured 36.3% 21.4%
Wrong category or variant% of products, lower is better 37 · 33 of 37 measured 11.3% 5.4%
Wrong category or variant · sales brief% of products, lower is better 37 · 33 of 37 measured 13.1% 5.4%
DeckBench measured by Spring Prompt · 14 results
Caveat dropped% of tasks, lower is better 1 · 16 of 37 measured 0.0% 16.7%
Clean layout% of tasks, higher is better 10 · 31 of 37 measured 66.7% 0.0%
Design quality% of the maximum, higher is better 34 · 26 of 37 measured 43.7% 57.3%
Draft figure quoted% of tasks, lower is better 30 · 21 of 37 measured 50.0% 33.3%
Findings missing% of tasks, lower is better 1 · 25 of 37 measured 0.0% 16.7%
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. 35 · 25 of 37 measured 25.2% 42.9%
Layout defects% of tasks, lower is better 10 · 31 of 37 measured 33.3% 100.0%
Misleading metric used% of tasks, lower is better 1 · 31 of 37 measured 0.0% 16.7%
Presentable decks% of tasks, higher is better 17 · 17 of 37 measured 0.0% 0.0%
Recommendation late or wrong% of tasks, lower is better 28 · 31 of 37 measured 50.0% 66.7%
Slides needing work% of tasks, lower is better 20 · 12 of 37 measured 50.0% 33.3%
Time per deckseconds, lower is better 2 · 36 of 37 measured 25.0 718
Unsupported claims% of tasks, lower is better 26 · 35 of 37 measured 33.3% 66.7%
Unsupported numbers% of tasks, lower is better 1 · 28 of 37 measured 0.0% 16.7%
ROASBench measured by Spring Prompt · 6 results
Consistency scorescore out of 100, higher is better 35 · 37 of 37 measured 8.7 7.7
Contribution profitsimulated US dollars, higher is better 36 · 27 of 37 measured −$159,882 $364,687
Cost of a runUS dollars, lower is better 14 · 7 of 37 measured $0.15 $0.0351
Months over budgetmonths of 12, lower is better 37 · 1 of 37 measured 3 0
Planning scorescore out of 100, higher is better 3 · 23 of 37 measured 55.2 54.7
Return on ad spendprofit per $1 spent, higher is better 36 · 27 of 37 measured −$0.17 $0.31
SurveyBench measured by Spring Prompt · 17 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. 26 · 25 of 28 measured 26.2% 28.8%
Leading or double-barrelled questions% of tasks, lower is better 22 · 12 of 28 measured 55.6% 33.3%
Mistaken next steps% of tasks, lower is better 24 · 28 of 28 measured 77.8% 88.9%
Numbers right% of answers, higher is better 25 · 1 of 28 measured 75.5% 100.0%
Objectives unanswerable% of tasks, lower is better 18 · 24 of 28 measured 11.1% 22.2%
Precision ignored% of tasks, lower is better 26 · 1 of 28 measured 100.0% 0.0%
Research scorepoints of 100, higher is better 21 · 27 of 28 measured 64.2 38.5
Sound research% of tasks, higher is better 17 · 17 of 28 measured 0.0% 0.0%
Stakeholder's view presumed% of tasks, lower is better 26 · 12 of 28 measured 33.3% 11.1%
Survey plan scorepoints of 100, higher is better 25 · 19 of 28 measured 81.5 89.8
Survey structure% of tasks, lower is better 3 · 22 of 28 measured 11.1% 33.3%
Time per taskseconds, lower is better 2 · 27 of 28 measured 47.5 1,000
Traps handled% of traps, higher is better 27 · 20 of 28 measured 59.4% 80.0%
Traps missed% of tasks, lower is better 25 · 25 of 28 measured 100.0% 100.0%
Unsupported findings% of tasks, lower is better 24 · 24 of 28 measured 100.0% 100.0%
Wrong numbers% of tasks, lower is better 26 · 22 of 28 measured 100.0% 77.8%
Wrong recommendation% of tasks, lower is better 19 · 24 of 28 measured 44.4% 77.8%

Quick answers

Is Mistral Medium 3.5 better than Inkling 2 Mini?

Pick Inkling 2 Mini: it scores higher (Spring Prompt overall 21 against 15) and costs less.

Which is cheaper, Mistral Medium 3.5 or Inkling 2 Mini?

Inkling 2 Mini is cheaper to run: $0.001 against $0.008 per product listing on CatalogBench (5.4 times less).

Which is better for product listings, Mistral Medium 3.5 or Inkling 2 Mini?

Mistral Medium 3.5 ranks 26 of 36 models on sale for turning a sparse product feed and photos into listings that can go live, against 28 for Inkling 2 Mini.

Which is better for decks from an analysis, Mistral Medium 3.5 or Inkling 2 Mini?

Inkling 2 Mini ranks 29 of 36 models on sale for turning a finished analysis into a deck you could present as it is, against 35 for Mistral Medium 3.5.

Which is better for user surveys, Mistral Medium 3.5 or Inkling 2 Mini?

Mistral Medium 3.5 ranks 24 of 27 models on sale for planning a user survey and reading its results without being misled, against 26 for Inkling 2 Mini.

Which is better for marketing planning, Mistral Medium 3.5 or Inkling 2 Mini?

Inkling 2 Mini ranks 29 of 36 models on sale for planning a year of ad spend without overspending, against 36 for Mistral Medium 3.5.

Which is better for speed under pressure, Mistral Medium 3.5 or Inkling 2 Mini?

Mistral Medium 3.5 ranks 13 of 38 models on sale for good decisions against a real clock (fast chess), against 39 for Inkling 2 Mini.

Which has the bigger context window, Mistral Medium 3.5 or Inkling 2 Mini?

Inkling 2 Mini: 1,048,576 tokens, against 262,144 for Mistral Medium 3.5.

Run Mistral Medium 3.5 and Inkling 2 Mini 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.