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 · Proprietary · rank 31 · bench-align-v5
Inkling 2 Small is built for coding and agent work. It ranks in the top quarter for multimodal tasks (top 9% of 136 scored models) and agent work (top 13% of 127 scored models), by BenchLM's published scores. It accepts text and images and produces text.
Inkling 2 Small by Thinking Machines Lab scores 66.7 overall, costs $0.30 per 1M input tokens and $1.00 per 1M output tokens, and carries a context window of 1M tokens — observed 2026-11-21.
Every published price component, including cache reads and writes.
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.
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.
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.
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 | 209 tok/s | 2026-11-21 | 2026-11-21 |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| GPQA Graduate-Level Google-Proof Q&A | 93 | #14 | 448 questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| GPQA-D GPQA Diamond | 93.0 | #15 | Graduate-level science questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| HLE Humanity's Last Exam | 53.2 | #19 | Expert-level questions | Frontier expert level | Humanity's Last Exam |
| Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index | 30.6 | #58 | Cross-benchmark intelligence index | Display-only external reference | Artificial Analysis |
| AA-HLE Artificial Analysis Humanity's Last Exam | 44.1 | #29 | Expert-level questions | Frontier expert reasoning | Artificial Analysis Humanity's Last Exam Benchmark Leaderboard |
| AA-Omniscience Index Artificial Analysis Omniscience Index | 16.1 | #42 | Knowledge questions | Broad factual knowledge | AA-Omniscience: Knowledge and Hallucination Benchmark |
| AA-Omniscience Accuracy Artificial Analysis Omniscience Accuracy | 33.8 | #82 | Knowledge questions | Broad knowledge | Artificial Analysis model benchmarks |
| AA-Omniscience Hallucination Rate Artificial Analysis Omniscience Hallucination Rate | 26.7 | #199 | Knowledge questions | Factuality | Artificial Analysis model benchmarks |
| HLE w/o tools Humanity's Last Exam without tools | 44.1 | #20 | 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 | 85.1 | #19 | Terminal-based software-agent tasks | Professional software engineering | Terminal-Bench |
| SWE-bench Verified Software Engineering Benchmark Verified | 82.2 | #12 | 500 verified issues | Professional software engineering | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? |
| SWE Multilingual SWE Multilingual | 81.2 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| HMMT Feb 2026 Harvard-MIT Mathematics Tournament February 2026 | 95.8 | #2 | Competition math problems | Olympiad-style mathematics | Qwen3.6 launch benchmarks |
| Apex Apex | 63.5 | #3 | Advanced mathematical reasoning | Frontier math reasoning | DeepSeek-V4 Technical Report |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| ARC-AGI-1 ARC-AGI-1 Semi-Private Evaluation | 96.50 | #11 | Semi-private ARC-AGI-1 evaluation set | Abstract visual reasoning | ARC Prize leaderboard |
| LongBench v2 LongBench v2 | 70.6 | #1 | Long-context tasks | Hard long-context | LongBench v2 |
| ARC-AGI-2 Abstraction and Reasoning Corpus for AGI v2 | 81.666667 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA Briefcase Artificial Analysis Briefcase | 1527 | #23 | Professional knowledge-work tasks | Professional work | Artificial Analysis Briefcase Benchmark Leaderboard |
| AA AutomationBench Artificial Analysis AutomationBench | 45.6 | #40 | Business-process automation tasks | Agentic automation | Artificial Analysis AutomationBench Benchmark Leaderboard |
| AA EnterpriseOps-Gym Artificial Analysis EnterpriseOps-Gym | 46.8 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Chartography (no tools) Chartography without tools | 36.7 | #10 | 100 specialized chart types | Professional chart reasoning | Chartography |
| Chartography (tools) Chartography with image and code tools | 80.4 | #8 | 100 specialized chart types | Professional chart reasoning | Chartography |
| MMMU-Pro Massive Multi-discipline Multimodal Understanding Pro | 81.6 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| PMPP-Hard PMPP-Hard GPU kernel optimization (AUC, 0-10 hours) | 494.27 | #1 | 69 hard GPU kernel-optimization tasks | Long-horizon GPU kernel optimization | Introducing Inkling 2 Small |
| ScreenSpot Pro w/ Python ScreenSpot Pro with Python | 91.5 | #1 | 1,581 GUI-grounding instructions on professional screenshots | High-resolution GUI grounding with tools | Introducing Inkling 2 Small |
| ComplexConstraints ComplexConstraints (Surge) | 44.5 |
| #9 |
| Multilingual software-engineering tasks |
| Professional software engineering |
| MiniMax M2.7: Early Echoes of Self-Evolution |
| FrontierSWE v2 FrontierSWE v2 | 25.1 | #19 | 34 ultra-long-horizon engineering and research tasks | Ultra-long-horizon frontier software engineering | FrontierSWE v2 |
| AA-SciCode Artificial Analysis SciCode | 52.3 | #42 | Scientific coding subproblems | Scientific programming | Artificial Analysis SciCode Benchmark Leaderboard |
| DeepSWE DeepSWE | 72.1 | #20 | 113 software engineering tasks across 91 repositories and 5 languages | Long-horizon software engineering | DeepSWE benchmark blog |
| FrontierCode 2.0 FrontierCode 2.0 | 22.6 | #24 | 132 private software-engineering tasks | Frontier coding-agent quality | FrontierCode 2.0 |
| PostTrainBench PostTrainBench v1.2 | 31.9 | #13 | Post-training four base models on six benchmarks | Autonomous ML engineering | PostTrainBench |
| Visual pattern completion and abstract reasoning |
| Expert-level — hardest public reasoning benchmark |
| ARC-AGI-2: A Harder General Intelligence Benchmark |
| ARC-AGI-3 Abstraction and Reasoning Corpus for AGI v3 | 0.29737 | #21 | Interactive game-like tasks with hidden rules | Frontier agentic reasoning | ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence |
| AA-LCR Artificial Analysis Long Context Reasoning | 83.3 | #13 | Long-context reasoning tasks | Long-context reasoning | Artificial Analysis model benchmarks |
| MLCR-AA Medical Long Context Reasoning (MLCR-AA) | 20.6 | #27 | Long, fragmented medical-record reasoning | Long-context medical reasoning | Medical Long Context Reasoning (MLCR-AA) |
| ARC-AGI-3 Provider Adapter ARC-AGI-3 Semi-Private: Provider Adapter harness | 2.55 | #12 | — | — | ARC Prize verified ARC-AGI-3 results |
| #10 |
| Enterprise operations workflows |
| Enterprise agent operations |
| Artificial Analysis EnterpriseOps-Gym Benchmark Leaderboard |
| AA Tau3 Banking Artificial Analysis Tau3-Banking | 40.4 | #14 | Banking tool-use workflows | Agentic banking workflows | Artificial Analysis Tau3-Banking Benchmark Leaderboard |
| Terminal-Bench 2.1 Terminal-Bench 2.1 | 85.1 | #20 | Terminal-based software-agent tasks | Professional software engineering | Terminal-Bench |
| BrowseComp BrowseComp | 90.2 | #9 | Research questions requiring browsing | Hard web research | BrowseComp |
| GDPval-AA GDPval-AA | 1630 | #23 | Agentic real-world work tasks | Professional agentic workflows | DeepSeek-V4 Technical Report |
| AA Terminal-Bench 4.0 Artificial Analysis Terminal-Bench v4.0 | 13.1 | #37 | Terminal-based agent tasks | Agentic software engineering | Artificial Analysis Terminal-Bench v4.0 Benchmark Leaderboard |
| GDP.pdf Artificial Analysis GDP.pdf | 25.2 | #18 | Professional document-production tasks | Professional knowledge work | GDP.pdf Benchmark Leaderboard |
| AA-AnalystAgent Artificial Analysis AnalystAgent | 51.2 | #10 | Spreadsheet and document analysis questions | Business and data analysis | AA-AnalystAgent Benchmark Leaderboard |
| JobBench JobBench | 70.5 | #2 | 130 tasks across 35 occupations | Professional multi-source workflows | JobBench: Aligning Agent Work With Human Will |
| MCP Atlas MCP Atlas | 85 | #5 | Tool-integrated agent tasks | Advanced tool use | Introducing GPT-5.4 mini and nano |
| DeepSearchQA DeepSearchQA | 90.9 | #9 | Agentic browsing and list-answer questions | Agentic web research | Muse Spark Eval Methodology |
| #7 |
| Multimodal academic reasoning |
| Frontier multimodal |
| MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark |
| AA-MMMU-Pro Artificial Analysis MMMU-Pro | 81.6 | #16 | Multimodal academic reasoning | Frontier multimodal | Artificial Analysis MMMU-Pro Benchmark Leaderboard |
| BabyVision w/ Python BabyVision with Python | 93.7 | #3 | Visual perception tasks with Python | Fine-grained visual perception | Kimi K3: Open Frontier Intelligence |
| PerceptionBench PerceptionBench (Internal) | 64.7 | #2 | Internal atomic visual-perception tasks | Fine-grained visual perception | Kimi K3: Open Frontier Intelligence |
| MathVision MathVision | 92.8 | #3 | Visually grounded math problems | Advanced multimodal mathematics | Qwen3.6 launch benchmarks |
| ScreenSpot Pro ScreenSpot Pro | 83.8 | #7 | 1,581 grounding instructions | Professional GUI grounding | ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use |
| CharXiv CharXiv Reasoning | 87.7 | #18 | Scientific chart reasoning | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
| CharXiv w/o tools CharXiv Reasoning without tools | 85.6 | #5 | Scientific chart reasoning (tool-free) | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
| BabyVision BabyVision | 68.6 | #5 | Visual perception tasks | Fine-grained visual perception | Muse Spark 1.1 Evaluation Report |
| #1 |
| Prompts with multiple interacting constraints |
| Constrained instruction following |
| Introducing Inkling 2 Small |
| EnterpriseBench-CoreCraft EnterpriseBench: CoreCraft (Surge) | 71.8 | #1 | Enterprise operations tasks in a simulated company | Multi-step enterprise tool use | Introducing Inkling 2 Small |
| Diligence Bench DiligenceBench (Thoughtful), best harness | 58.1 | #1 | Financial due-diligence research tasks | Agentic financial analysis | Introducing Inkling 2 Small |
| DAYJOB (Finance) DAYJOB, finance track | 14.3 | #1 | Professional finance job tasks | Long-horizon finance work | Introducing Inkling 2 Small |
| StrongREJECT StrongREJECT (no jailbreak) | 97.8 | #1 | Forbidden requests across harm categories | Harmful-request refusal | A StrongREJECT for Empty Jailbreaks |
| PatchEval PatchEval | 79.6 | #1 | Real-world vulnerability patching tasks | Security vulnerability repair | Introducing Inkling 2 Small |
| FrontierPhysics End-to-End Physics Research Benchmark for AI Agents | 3.0 | #23 | 56 research projects from 42 physicists across nine fields, three trials each (168 rollouts per setup) | Open-ended research projects that took their authors about 270 days on average | FrontierPhysics: Evaluating Agents for End-to-End Physics Research (arXiv, November 4, 2026) |
| HoneyBench HoneyBench v0.2 (Goodhart Labs) | 52.6 | #17 | 20 private honeypot tasks across 7 domains with exploitable evaluators | Agentic alignment and specification gaming | HoneyBench v0.2 |