Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Thinking Machines Lab · Open Weight · rank 69 · bench-align-v5
Inkling 2 Mini is a general-purpose language model.
Inkling 2 Mini by Thinking Machines Lab scores 56 overall, costs $0.15 per 1M input tokens and $0.50 per 1M output tokens, and carries a context window of 1M tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Each axis shows the ranking source’s score and, beside it, BenchLM’s separate verified score for that axis. The two are published side by side and are not interchangeable.
Every published price component, including cache reads and writes.
Lifecycle events the source associates with this model, plus what this profile does not claim.
That is not evidence the model has no lifecycle plan — only that this source published none.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: ObservedUnavailable
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | Unavailable | Unavailable | Unavailable | |
| Throughput | Unavailable | Unavailable | Unavailable |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| GPQA Graduate-Level Google-Proof Q&A | 86.6 | #50 | 448 questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| GPQA-D GPQA Diamond | 86.6 | #56 | Graduate-level science questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| HLE w/o tools Humanity's Last Exam without tools | 30.7 | #43 | Expert-level questions | Frontier expert level | Introducing GPT-5.4 mini and nano |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 2.1 Terminal-Bench 2.1 | 71.6 | #48 | Terminal-based software-agent tasks | Professional software engineering | Terminal-Bench |
| SWE-bench Verified Software Engineering Benchmark Verified | 79.2 | #26 | 500 verified issues | Professional software engineering | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? |
| SWE Multilingual SWE Multilingual | 73.7 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Apex Apex | 34.4 | #8 | Advanced mathematical reasoning | Frontier math reasoning | DeepSeek-V4 Technical Report |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| LongBench v2 LongBench v2 | 66.9 | #5 | Long-context tasks | Hard long-context | LongBench v2 |
| MRCR v2 256K-512K OpenAI MRCR v2 8-needle 256K-512K | 46.4 | #3 | 100 eight-needle retrieval examples | Very long-context retrieval | Muse Spark 1.3 Evaluation Methodology |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 4.0 Terminal-Bench 4.0 | 5.60 | #29 | 66 professional computer-work tasks | Frontier autonomous knowledge work | Terminal-Bench 4.0 |
| Terminal-Bench-Science 0.1 Terminal-Bench-Science 0.1 | 1.4 | #15 | 70 expert-curated scientific research workflows | Frontier scientific research workflows | Terminal-Bench-Science 0.1 |
| Terminal-Bench 2.1 Terminal-Bench 2.1 | 71.6 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Chartography (no tools) Chartography without tools | 22.1 | #12 | 100 specialized chart types | Professional chart reasoning | Chartography |
| Chartography (tools) Chartography with image and code tools | 72.5 | #10 | 100 specialized chart types | Professional chart reasoning | Chartography |
| MMMU-Pro Massive Multi-discipline Multimodal Understanding Pro | 74.9 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| MultiChallenge (provider reports) MultiChallenge: provider-reported runs | 50.00 | #4 | — | — | MultiChallenge |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| PMPP-Hard PMPP-Hard GPU kernel optimization (AUC, 0-10 hours) | 455.82 | #2 | 69 hard GPU kernel-optimization tasks | Long-horizon GPU kernel optimization | Introducing Inkling 2 Small |
| ScreenSpot Pro w/ Python ScreenSpot Pro with Python | 87.7 | #2 | 1,581 GUI-grounding instructions on professional screenshots | High-resolution GUI grounding with tools | Introducing Inkling 2 Small |
| ComplexConstraints ComplexConstraints (Surge) | 33.4 |
| #31 |
| Multilingual software-engineering tasks |
| Professional software engineering |
| MiniMax M2.7: Early Echoes of Self-Evolution |
| DeepSWE DeepSWE | 48.2 | #57 | 113 software engineering tasks across 91 repositories and 5 languages | Long-horizon software engineering | DeepSWE benchmark blog |
| FrontierCode 2.0 FrontierCode 2.0 | 15.1 | #31 | 132 private software-engineering tasks | Frontier coding-agent quality | FrontierCode 2.0 |
| #52 |
| Terminal-based software-agent tasks |
| Professional software engineering |
| Terminal-Bench |
| BrowseComp BrowseComp | 79 | #33 | Research questions requiring browsing | Hard web research | BrowseComp |
| JobBench JobBench | 49.6 | #17 | 130 tasks across 35 occupations | Professional multi-source workflows | JobBench: Aligning Agent Work With Human Will |
| MCP Atlas MCP Atlas | 82.7 | #12 | Tool-integrated agent tasks | Advanced tool use | Introducing GPT-5.4 mini and nano |
| DeepSearchQA DeepSearchQA | 85.3 | #12 | Agentic browsing and list-answer questions | Agentic web research | Muse Spark Eval Methodology |
| EnterpriseOps-Gym EnterpriseOps-Gym (provider runs) | 42.00 | #1 | Enterprise workflow tasks | Stateful enterprise tool use | EnterpriseOps-Gym: Environments and Evaluations for Stateful Agentic Planning and Tool Use in Enterprise Settings |
| #41 |
| Multimodal academic reasoning |
| Frontier multimodal |
| MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark |
| BabyVision w/ Python BabyVision with Python | 88.8 | #7 | Visual perception tasks with Python | Fine-grained visual perception | Kimi K3: Open Frontier Intelligence |
| ZeroBench w/ Python ZeroBench_main with Python | 34.0 | #7 | Visual reasoning questions with Python | Tool-augmented visual reasoning | Kimi K3: Open Frontier Intelligence |
| PerceptionBench PerceptionBench (Internal) | 57.5 | #5 | Internal atomic visual-perception tasks | Fine-grained visual perception | Kimi K3: Open Frontier Intelligence |
| MathVision MathVision | 89.0 | #10 | Visually grounded math problems | Advanced multimodal mathematics | Qwen3.6 launch benchmarks |
| ScreenSpot Pro ScreenSpot Pro | 76.5 | #10 | 1,581 grounding instructions | Professional GUI grounding | ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use |
| ZeroBench ZeroBench | 17.0 | #10 | 100 visual reasoning questions | Tool-augmented visual reasoning | Muse Spark Eval Methodology |
| CharXiv CharXiv Reasoning | 88.3 | #16 | Scientific chart reasoning | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
| CharXiv w/o tools CharXiv Reasoning without tools | 80.6 | #14 | Scientific chart reasoning (tool-free) | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
| BabyVision BabyVision | 53.4 | #8 | Visual perception tasks | Fine-grained visual perception | Muse Spark 1.1 Evaluation Report |
| OmniDocBench 1.6 OmniDocBench v1.6 | 92.8 | #1 | Document parsing and extraction | Grounded document reasoning | OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations |
| #2 |
| Prompts with multiple interacting constraints |
| Constrained instruction following |
| Introducing Inkling 2 Small |
| Diligence Bench DiligenceBench (Thoughtful), best harness | 48.6 | #2 | Financial due-diligence research tasks | Agentic financial analysis | Introducing Inkling 2 Small |
| StrongREJECT StrongREJECT (no jailbreak) | 97.8 | #1 | Forbidden requests across harm categories | Harmful-request refusal | A StrongREJECT for Empty Jailbreaks |
| BVB BVB (Vision Interpretation and Creation) | 68.4 | #1 | Vision interpretation and creation tasks (as reported by Thinking Machines Lab) | Visual interpretation and creation | Introducing Inkling 2 Mini |
| Intelligence per Watt Intelligence per Watt (local AI energy efficiency) | 0.703 | #1 | Local LLM query efficiency benchmark suite | Energy-efficient local inference | Intelligence per Watt: Measuring Intelligence Efficiency of Local AI |
| CritPt v2 (Full) CritPt v2 Full set (research-level physics) | 31.4 | #1 | Composite unpublished research-level physics challenges (Full set) | Frontier research-level physics reasoning | Introducing Inkling 2 Mini |
| CritPt v2 (Hard) CritPt v2 Hard subset (research-level physics) | 17.0 | #1 | Hard subset of CritPt v2 physics challenges | Hard frontier physics reasoning | Introducing Inkling 2 Mini |
| GraphWalks BFS 256K GraphWalks BFS at 256K context | 69.7 | #1 | Multi-hop BFS graph traversal problems at 256K context | Long-context multi-hop graph reasoning | Introducing Inkling 2 Mini |
| FORTRESS (adversarial, Public) FORTRESS Public adversarial safety split | 89.5 | #1 | Public adversarial safety prompts from FORTRESS | Adversarial safety and refusal | Introducing Inkling 2 Mini |
| FORTRESS (benign, Public) FORTRESS Public benign over-refusal split | 96.4 | #1 | Public benign safety-adjacent prompts from FORTRESS | Calibration against over-refusal | Introducing Inkling 2 Mini |