Models / Inkling 2 Mini

Thinking Machines

Inkling 2 Mini

Inkling 2 Mini trails most of the field on what it has been measured on; worth a look for simple, high-volume work where cost matters most.

Results 21 Nov 2026 · 80 results · 4 sources

Test Inkling 2 Mini on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.09 in · $0.36 out
OpenRouter, 10 Nov 2026 · compare prices
Speed
Not measured
Intelligence Index
Not measured
Spring Prompt overall
37th of 40 score 21 of 100 · how it works
Spring Prompt Score
19 of 100 35th of 36 · 4 of 5 benchmarks · our own tests only
Developer
Thinking Machines, based in the US
Weights published
Released
20 Nov 2026
1,048,576 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today – 36th of 39Claude Fable 5.2 $0.16 blended, 14th percentilecheapest: Mistral Nemo –
Models priced under $0.25 per million tokens – 5th of 5Muse Ember sets this group –
Models from developers based in the US – 24th of 25Claude Fable 5.2 $0.16 blended, 15th percentilecheapest: Gemma 3 4B –
Open-weights models – 9th of 9Kimi K3.1 Pro $0.16 blended, 23rd percentilecheapest: Mistral Nemo –

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

Measured cost per task

API cost recorded in our benchmark runs, excluding judging.

TaskBenchmarkCost per task
Product listingCatalogBench≈ $0.001
DeckDeckBench≈ $0.022
Survey plan and analysisSurveyBench≈ $0.017

Cheapest model that passes

On our benchmarks that define a pass for each task: what one passing answer cost in our run, failed attempts included, and whether a cheaper model passed.

  • CatalogBench · cost per 1,000 accepted listings
    Inkling 2 Mini: $3.16 (79 of 168 attempts), 3 of 28. Cheapest: GPT-6 Luna at $0.70 (114 of 168 attempts). It passed at least as often.
  • DeckBench · cost per presentable deck
    Inkling 2 Mini: no presentable decks (0 of 6 tasks). Cheapest: GPT-6.2 Sol at $0.09 (5 of 6 tasks).
  • SurveyBench · cost per sound survey task
    Inkling 2 Mini: no sound survey tasks (0 of 9 tasks). Cheapest: GPT-6 Luna at $0.03 (3 of 9 tasks).

Cost per attempt divided by the share that passed, at the run date's prices, excluding judging. Each model at its cheapest setting per passing answer. With a handful of tasks, a pass or two either way is within chance. No estimate where a run had missing outputs, because their cost may not have been recorded.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced under $0.25 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live 28th of 36 available · 4th of 5 at its price
BenchmarkInkling 2 MiniRank among availableBest available
CatalogBenchCatalogBench: reliably publish-ready 23.2% 28th of 36 Gemini 4 Neon 78.6%
CatalogBenchCatalogBench (sales brief): reliably publish-ready 5.4% 25th of 36 GPT-6.2 Sol 75.0%
Decks from an analysisTurning a finished analysis into a deck you could present as it is 29th of 36 available · 5th of 5 at its price
BenchmarkInkling 2 MiniRank among availableBest available
DeckBenchDeckBench: deck rating 792 29th of 36 Qwen4-Max 1,569
User surveysPlanning a user survey and reading its results without being misled 26th of 27 available · 5th of 5 at its price
BenchmarkInkling 2 MiniRank among availableBest available
SurveyBenchSurveyBench: analysis rating 567 26th of 27 Claude Fable 5.2 1,679
Marketing planningPlanning a year of ad spend without overspending 29th of 36 available · 4th of 5 at its price
BenchmarkInkling 2 MiniRank among availableBest available
ROASBenchROASBench: overall score 29.9 29th of 36 Claude Fable 5.2 61.3
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls PendingReleased after these sources last updated
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
  • Artificial Analysis: pending. Its latest results are dated 9 Nov 2026, before this model was released.
  • Vending-Bench 2: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects PendingReleased after these sources last updated
  • APEX-Agents: pending. Its latest results are dated 1 Oct 2026, before this model was released.
  • GDP.pdf: pending. Its latest results are dated 1 Oct 2026, before this model was released.
  • Remote Labor Index: pending. Its latest results are dated 1 Oct 2026, before this model was released.
Reasoning and knowledgeHard questions across science, maths and general knowledge PendingReleased after these sources last updated
  • Artificial Analysis: pending. Its latest results are dated 20 Oct 2026, before this model was released.
WritingWhat people prefer in blind comparisons, and judged writing quality PendingReleased after these sources last updated
  • Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
  • UGI Leaderboard: pending. Its latest results are dated 2 Oct 2026, before this model was released.
Sticking to the factsSummarising without inventing things, and factual answers PendingReleased after these sources last updated
  • Vectara Hallucination Leaderboard: pending. Its latest results are dated 22 Sep 2026, before this model was released.
  • Arena (formerly LMArena): pending. Its latest results are dated 2 Oct 2026, before this model was released.
  • SimpleQA Verified (Epoch AI): pending. Its latest results are dated 29 Sep 2026, before this model was released.
Speed under pressureGood decisions against a real clock (fast chess) Too slow for the clock
  • BulletBench: too slow for Blitz 3+2: lost its first 4 games on time, at a median 11.9 s a move; too slow for Bullet 60s: lost its first 4 games on time, at a median 3.8 s a move; too slow for Lightning 10+1: lost its first 4 games on time, at a median 6.8 s a move.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Inkling 2 MiniMuse Ember
highest-scoring overall at under $0.25 per million tokens
Claude Fable 5.2
highest-scoring overall available today
Gemini 3.5 Flash-Lite
just above overall
Mistral Medium 3.5
just below overall
Price per million tokens $0.16 $0.20$16.00$0.85$3.00
Tokens a second – ––398157
Intelligence Index – ––22.214.2
Spring Prompt overall 21 62882415
Rank among available models, by use case
Product listings 28th of 36 13th9th29th26th
Decks from an analysis 29th of 36 18th4th33rd35th
User surveys 26th of 27 8th1st21st24th
Marketing planning 29th of 36 10th1st32nd36th
Agents and tool use pending ––81st100th
Professional work pending ––46th–
Reasoning and knowledge pending ––98th123rd
Writing pending ––49th85th
Sticking to the facts pending ––61st98th
Speed under pressure 39th of 39 24th10th2nd13th

Shaded figures are better than Inkling 2 Mini's. A dash means no result.

Research and examples

In the playground: our pick in 0 of 2 tests, 86% of checks passed →

Where it has been measured

Among models available today, in the bottom quarter for decks from an analysis, user surveys, marketing planning and speed under pressure. Results for agents and tool use, professional work, reasoning and knowledge, writing, and sticking to the facts are pending: those sources last updated before this model was released.

Not timed: Artificial Analysis, our source for speed, has not tested it

Every result

80 published results from 4 sources, each in the source's own units, with the configuration that produced it.

CatalogBench · measured by Spring Prompt · 32 results
MeasureValueRankConfigurationDated
CatalogBench (sales brief): channel rules broken% of products, lower is better 3.6% 36 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): claims to check% of products, lower is better 1.8% 26 of 36 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): content quality% of checks, higher is better 94.1% 34 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 1.2% 31 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): not findable% of products, lower is better 3.6% 17 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 27.4% 22 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 5.4% 25 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 61.9% 23 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 2.01 23 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 21.4% 30 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 5.4% 34 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: channel compliance% of products, higher is better 98.8% 35 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: channel rules broken% of products, lower is better 3.0% 36 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: claims to check% of products, lower is better 0.0% 1 of 36 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: content quality% of checks, higher is better 95.3% 33 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: cost per productUS dollars, lower is better $0.0015 6 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: decision accuracy% of decisions, higher is better 95.3% 30 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: failed outputs% of products, lower is better 0.0% 1 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: field accuracy% of missing fields, higher is better 93.7% 18 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: invented values% of filled values, lower is better 3.8% 33 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: not findable% of products, lower is better 4.2% 18 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: output tokens per producttokens, lower is better 3,125 12 of 16 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: publish-ready listings% of products, higher is better 47.0% 26 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: reliably publish-ready% of products, higher is better 23.2% 28 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: time per productseconds, lower is better 40.7 31 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: unsupported claims% of products, lower is better 27.4% 27 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: unsupported claimsclaims per product, lower is better 1.24 34 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: wrong attributes% of products, lower is better 24.4% 34 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
CatalogBench: wrong category or variant% of products, lower is better 5.4% 35 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026

Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.

DeckBench · measured by Spring Prompt · 18 results
MeasureValueRankConfigurationDated
DeckBench: accurate decks% of tasks, higher is better 0.0% 32 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: caveat dropped% of tasks, lower is better 16.7% 16 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: clean layout% of tasks, higher is better 0.0% 31 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: cost per deckUS dollars, lower is better $0.0223 10 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: deck ratingrating, higher is better 792
669–907
30 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: design quality% of the maximum, higher is better 57.3% 26 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: draft figure quoted% of tasks, lower is better 33.3% 21 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: findings missing% of tasks, lower is better 16.7% 25 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: head-to-head win rate% of comparisons, higher is better 42.9% 26 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: layout defects% of tasks, lower is better 100.0% 31 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: misleading metric used% of tasks, lower is better 16.7% 31 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: output tokens per decktokens, lower is better 51,588 16 of 17 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: presentable decks% of tasks, higher is better 0.0% 17 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: recommendation late or wrong% of tasks, lower is better 66.7% 32 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: slides needing work% of tasks, lower is better 33.3% 12 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: time per deckseconds, lower is better 718 37 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: unsupported claims% of tasks, lower is better 66.7% 36 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026
DeckBench: unsupported numbers% of tasks, lower is better 16.7% 28 of 38 Inkling 2 Mini
provider default reasoning
21 Nov 2026

Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.

ROASBench · measured by Spring Prompt · 9 results
MeasureValueRankConfigurationDated
ROASBench: audience scorescore out of 100, higher is better 54.0 29 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: business scorescore out of 100, higher is better 24.1 30 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: consistency scorescore out of 100, higher is better 7.7 39 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: contribution profitsimulated US dollars, higher is better $364,687 28 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: cost of a runUS dollars, lower is better $0.0351 7 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: months over budgetmonths of 12, lower is better 0 1 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: overall scorescore out of 100, higher is better 29.9 30 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: planning scorescore out of 100, higher is better 54.7 23 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.31 28 of 39 Inkling 2 Mini
provider default reasoning
21 Nov 2026

Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

SurveyBench · measured by Spring Prompt · 21 results
MeasureValueRankConfigurationDated
SurveyBench: analysis ratingrating, higher is better 567
385–664
27 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: analysis scorepoints of 100, higher is better 16.5 25 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: cost per taskUS dollars, lower is better $0.0168 4 of 28 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: head-to-head win rate% of comparisons, higher is better 28.8% 25 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: leading or double-barrelled questions% of tasks, lower is better 33.3% 12 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: mistaken next steps% of tasks, lower is better 88.9% 28 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: numbers right% of answers, higher is better 100.0% 1 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: objectives unanswerable% of tasks, lower is better 22.2% 24 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: output tokens per tasktokens, lower is better 43,704 13 of 18 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: precision ignored% of tasks, lower is better 0.0% 1 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: research scorepoints of 100, higher is better 38.5 27 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: sound research% of tasks, higher is better 0.0% 17 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: stakeholder's view presumed% of tasks, lower is better 11.1% 12 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: survey plan scorepoints of 100, higher is better 89.8 19 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: survey structure% of tasks, lower is better 33.3% 22 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: time per taskseconds, lower is better 1,000 27 of 28 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: traps handled% of traps, higher is better 80.0% 20 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: traps missed% of tasks, lower is better 100.0% 25 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: unsupported findings% of tasks, lower is better 100.0% 24 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: wrong numbers% of tasks, lower is better 77.8% 22 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026
SurveyBench: wrong recommendation% of tasks, lower is better 77.8% 24 of 29 Inkling 2 Mini
provider default reasoning
21 Nov 2026

Not shown: Not your customers or your data: invented organisations with simulated respondents, nine public tasks, one run per model per task at the provider's default reasoning setting.

Embed a score badge

A badge for its Spring Prompt Score and each of our benchmarks that ranked Inkling 2 Mini, from the current results. Paste the code into a page or README; it links back to where the result comes from.

Inkling 2 Mini is #35 of 36 on the Spring Prompt Score (19 of 100, from 4 of 5 of our benchmarks), measured by Spring Prompt Spring Prompt Score · 4 of 5 benchmarks
<a href="https://springprompt.com/score#thinkingmachines-inkling-2-mini"><img src="https://springprompt.com/badge/score/thinkingmachines-inkling-2-mini.svg" alt="Inkling 2 Mini is #35 of 36 on the Spring Prompt Score (19 of 100, from 4 of 5 of our benchmarks), measured by Spring Prompt" height="20"></a>
Inkling 2 Mini is #28 of 36 on CatalogBench (Reliably publish-ready, factual brief: 23.2%), measured by Spring Prompt CatalogBench · Reliably publish-ready, factual brief
<a href="https://springprompt.com/benchmarks/catalogbench?focus=thinkingmachines-inkling-2-mini"><img src="https://springprompt.com/badge/catalogbench/thinkingmachines-inkling-2-mini.svg" alt="Inkling 2 Mini is #28 of 36 on CatalogBench (Reliably publish-ready, factual brief: 23.2%), measured by Spring Prompt" height="20"></a>
Inkling 2 Mini is #29 of 37 on DeckBench (Deck rating: 792), measured by Spring Prompt DeckBench · Deck rating
<a href="https://springprompt.com/benchmarks/deckbench?focus=thinkingmachines-inkling-2-mini"><img src="https://springprompt.com/badge/deckbench/thinkingmachines-inkling-2-mini.svg" alt="Inkling 2 Mini is #29 of 37 on DeckBench (Deck rating: 792), measured by Spring Prompt" height="20"></a>
Inkling 2 Mini is #27 of 28 on SurveyBench (Analysis rating: 567), measured by Spring Prompt SurveyBench · Analysis rating
<a href="https://springprompt.com/benchmarks/surveybench?focus=thinkingmachines-inkling-2-mini"><img src="https://springprompt.com/badge/surveybench/thinkingmachines-inkling-2-mini.svg" alt="Inkling 2 Mini is #27 of 28 on SurveyBench (Analysis rating: 567), measured by Spring Prompt" height="20"></a>
Inkling 2 Mini is #29 of 37 on ROASBench (Score: 29.9), measured by Spring Prompt ROASBench · Score
<a href="https://springprompt.com/benchmarks/roasbench?focus=thinkingmachines-inkling-2-mini"><img src="https://springprompt.com/badge/roasbench/thinkingmachines-inkling-2-mini.svg" alt="Inkling 2 Mini is #29 of 37 on ROASBench (Score: 29.9), measured by Spring Prompt" height="20"></a>

Will Inkling 2 Mini work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.