- Home
- AI Models
- Thinking Machines Lab
- Inkling 2 Mini
Thinking Machines LabLanguage modelOpen weights
Inkling 2 Mini
Inkling 2 Mini has no overall index value yet because there are too few independent benchmark results.
Summary
Inkling 2 Mini is Thinking Machines Lab's open multimodal looped transformer with 26.4 billion parameters. Its middle 12 of 24 layers run three times. The model processes text and images, supports up to one million context tokens, and is also available through Tinker.
On this page
All benchmark results
Every published result with its source, test setup and comparison models.
Show all 24 published benchmark results
Page 1 of 2
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
AA-AnalystAgent: Overall (Show measurement, test conditions, and source)
- Source value
- 46.25
- Metric
- pass^5 reliability
- Unit
- %
- Benchmark version
- AA cohort: 80 questions
- Category
- agentic
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Harness: Artificial Analysis Stirrup. Agent: Stirrup analyst agent. Tool policy: 100-turn cap; Python sandbox with source files, URL fetch, image viewing, and answer submission. Attempts: pass^5, 5 independent attempts per question. Reasoning: reasoning enabled. Score display: 46.3%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-aa-analyst-agent:AA cohort: 80 questions:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.. Harness: Artificial Analysis Stirrup. Agent: Stirrup analyst agent. Tool policy: 100-turn cap; Python sandbox with source files, URL fetch, image viewing, and answer submission. Attempts: pass^5, 5 independent attempts per question. Reasoning: reasoning enabled
- Context
- Questions, source files, and reference answers are private. pass^5 measures reliability across all five attempts, not pass@1.
AutomationBench-AA: Overall (Show measurement, test conditions, and source)
- Source value
- 41.7962769721
- Metric
- Objectives completed without guardrail violations
- Unit
- %
- Benchmark version
- AA private 657-task split of AutomationBench v1.0.6
- Category
- agentic
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 41.8%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-automationbench-aa:AA private 657-task split of AutomationBench v1.0.6:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
CritPt v2 (Hard Subset): Overall (Show measurement, test conditions, and source)
- Source value
- 17
- Metric
- pass@1 accuracy
- Unit
- %
- Benchmark version
- v2 hard subset, November 2026 grading
- Category
- reasoning
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 17.0%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-critpt-v2-hard-subset:v2 hard subset, November 2026 grading:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
- Context
- This is the 40-question hard subset of CritPt v2, not the full 70-question CritPt v2 run.
FrontierCode: Overall (Show measurement, test conditions, and source)
- Source value
- 15.1037626263
- Metric
- Mean score over three attempts per task
- Unit
- %
- Benchmark version
- 2.0
- Category
- coding
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 15.1%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-frontiercode-2-0:2.0:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
GDP.pdf: Overall (Show measurement, test conditions, and source)
- Source value
- 16.8
- Metric
- All-pass rate
- Unit
- %
- Benchmark version
- unspecified
- Category
- reasoning
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 16.8%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-gdp-pdf:unspecified:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
HLE-Diamond: Overall (Show measurement, test conditions, and source)
- Source value
- 15.2
- Metric
- pass@1 accuracy
- Unit
- %
- Benchmark version
- Diamond
- Category
- general
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 15.2%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-hle-diamond:Diamond:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
AA-Omniscience Accuracy: Overall (Show measurement, test conditions, and source)
- Source value
- 20.5333333333
- Metric
- AA-Omniscience Accuracy
- Unit
- %
- Benchmark version
- AA Omniscience diagnostic
- Category
- knowledge
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard. Reasoning: reasoning enabled. Score display: 20.5%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-omniscience-accuracy:AA Omniscience diagnostic:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.. Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard. Reasoning: reasoning enabled
- Context
- This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.
AA-Omniscience Hallucination Rate: Overall (Show measurement, test conditions, and source)
- Source value
- 36.6191275168
- Metric
- AA-Omniscience Hallucination Rate
- Unit
- %
- Benchmark version
- AA Omniscience diagnostic
- Category
- knowledge
- Task
- overall
- Task label
- Overall
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard. Reasoning: reasoning enabled. Score display: 36.6%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-omniscience-hallucination-rate:AA Omniscience diagnostic:standalone:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).. Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard. Reasoning: reasoning enabled
- Context
- This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.
Terminal-Bench: Overall (Show measurement, test conditions, and source)
- Source value
- 5.5555555556
- Metric
- Task success
- Unit
- %
- Benchmark version
- 4.0
- Category
- coding
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 5.6%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-terminal-bench-4:4.0:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
Terminal-Bench-Science: Overall (Show measurement, test conditions, and source)
- Source value
- 1.4285714286
- Metric
- pass@1 task success rate
- Unit
- %
- Benchmark version
- 0.1.0
- Category
- reasoning
- Task
- overall
- Task label
- Overall
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-11-21
- Methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Source model ID: artificial-analysis/da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402/inkling-2-mini/inkling-2-mini. Source model name: Inkling 2 Mini. Provider reported by source: Thinking Machines. Model snapshot: Inkling 2 Mini. Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Score display: 1.4%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-11-21-84d3e03f2230. Source payload hash: 84d3e03f22302bc2eceba8784ccb68cb4374e95019533e3f2493be870ff91d2e. Parser version: artificial-analysis-rsc-models-v4. Imported at: 2026-11-21. Source cohort: artificial-analysis-terminal-bench-science:0.1.0:v5.0.1:reasoning:effort=unspecified:suffix=unspecified
- Comparison methodology
- Exact AA configuration: Inkling 2 Mini (release inkling-2-mini, row inkling-2-mini, UUID da3ba2f4-f4f1-43b6-95dc-8e9fa2a48402, reasoning=true, effort=not specified, suffix=not specified). This source capture does not provide an average reasoning-token count.. Reasoning: reasoning enabled
ARC-AGI-1 Semi-Private accuracy (Show measurement, test conditions, and source)
- Source value
- 84.5
- Metric
- accuracy on the semi-private evaluation split
- Unit
- %
- Benchmark version
- ARC-AGI-1, Semi-Private Eval
- Category
- reasoning
- Direction
- Higher is better
- Source type
- official-leaderboard
- Evaluator
- ARC Prize
- Status
- active
- Retrieved at
- 2026-11-22
- Methodology
- Official benchmark leaderboard. Verified ARC Prize result for Inkling 2 Mini on the semi-private evaluation split at xhigh reasoning.
- Context
- The verified score table does not specify the exact harness or the number of scored tasks for this run.
ARC-AGI-2 Semi-Private accuracy (Show measurement, test conditions, and source)
- Source value
- 49.3
- Metric
- accuracy on the semi-private evaluation split
- Unit
- %
- Benchmark version
- ARC-AGI-2, Semi-Private Eval
- Category
- reasoning
- Direction
- Higher is better
- Source type
- official-leaderboard
- Evaluator
- ARC Prize
- Status
- active
- Retrieved at
- 2026-11-22
- Methodology
- Official benchmark leaderboard. Verified ARC Prize result for Inkling 2 Mini on the semi-private evaluation split at xhigh reasoning.
- Context
- The verified score table does not specify the exact harness or the number of scored tasks for this run.
Page 1 of 2
Similar models
More models of the same kind as Inkling 2 Mini, chosen by model line, developer, size, price and release date.
- TML-d24-varloop-researchThinking Machines Lab · same developer
- Inkling 2 SmallThinking Machines Lab · same developer
- Inkling SmallThinking Machines Lab · same developer
- InklingThinking Machines Lab · same developer
- Arete-27BNous Research · similar size
- Holo4-27BH Company · similar size
- Gemma 4 26B A4BGoogle · similar size
- DiffusionGemma 26B A4B ITGoogle · similar size
All specifications
Architecture
- Architecture
- Multimodal autoregressive looped transformer with sparse mixture-of-experts layers.
- Source: Architecture
- Parameters
- 26.4 billion total and 3.4 billion active. Repeated layers bring the compute to 5.4 billion effective parameters per token.
- Source: Parameters
Input and output
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
- Access
- Open weights on Hugging Face, with API access through Tinker and third-party providers.
- Source: Access
- Shutdown
- None announced
- no source
Further details
Frequently asked questions about Inkling 2 Mini
How large is the context window of Inkling 2 Mini?
1 M tokens. Separate limits may apply to input and output.
Can I run Inkling 2 Mini locally?
Yes, the weights are open.
Stated by Thinking Machines Lab
- Open weights
- yes
- License
- Apache 2.0
- Parameters
- 26.4B
- active per token
- 3.4B
Sources and data date
Show all 18 sources
Benchmark sources at a glance
Which independent evaluators tested this model. This is coverage, not a quality ranking.
| Evaluator | Test series | Composite indices | Recorded active measurements |
|---|---|---|---|
| Artificial Analysis | 11 | 0 | 13 |
| ARC Prize | 0 | 0 | 2 |
Without vendor claims and historical or disputed results; those stay labelled in the single values. 2 measurements are not yet assigned to a test series.
| Type | Evidence and data date |
|---|---|
| Input | Documented modalities. Retrieved November 21, 2026. |
| Output | Documented modalities. Retrieved November 21, 2026. |
| Research status | Research date November 22, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not published | Parallel Search reasoning setting Note The model scorecard does not disclose a specific reasoning setting. |
| Not published | Parallel Search time per task Note The matching leaderboard row displays no time per task. |
| Additional source | Thinking Machines Inkling 2 Mini release (retrieved November 21, 2026) · Thinking Machines: Inkling 2 Mini release · Editorial description reviewed November 21, 2026 |
| Additional source | Thinking Machines Tinker pricing (retrieved November 21, 2026) |
| Additional source | Artificial Analysis (retrieved November 21, 2026)123456789101112 |
| Additional source | Parallel Search Capability Leaderboard (retrieved November 22, 2026) |
| Additional source | ARC Prize (retrieved November 22, 2026) |
| Additional source | Model metadata (retrieved November 22, 2026) |