Inkling-Small

Leaderboards

Where Inkling-Small sits among the top 20 in each category.

By task

Scored from votes on each kind of prompt alone, on the same 1,000-point scale.

Test results

Every test Inkling-Small has a published result on, out of 100%, with its rank among the models that took it.

• GPQA

  #32 / 256

  Graduate-level science questions written to be hard to look up.

  89.5%

  Self-reported

• SWE-Bench Verified

  #13 / 120

  Fixing real bugs in real GitHub projects, checked by the projects' own tests.

  80.2%

  Self-reported

• Humanity's Last Exam

  #53 / 108

  Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

  31.6%

  Self-reported · In the text score· In the vision score

• MMMU-Pro

  #36 / 73

  College-level questions that need both an image and text to answer.

  74.0%

  Self-reported · In the vision score

• BrowseComp

  #25 / 68

  Finding hard-to-locate facts by browsing the web.

  77.4%

  Self-reported · In the agents score

• SWE-Bench Pro

  #40 / 61

  Harder, longer software engineering tasks from real codebases.

  55.9%

  Self-reported · In the coding score

• CharXiv-R

  #37 / 60

  Reading and reasoning about charts from scientific papers.

  77.4%

  Self-reported

• Terminal-Bench 2.1

  #33 / 46

  Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal. It evaluates how well models handle real-world, end-to-end tasks autonomously, including compiling code, training models, setting up servers, system administration, data science workflows, and security tasks.

  64.7%

  Self-reported

• IFBench

  #2 / 44

  Instruction Following Benchmark evaluating model's ability to follow complex instructions

  82.2%

  Self-reported

• Toolathlon

  #16 / 42

  Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.

  54.4%

  Self-reported · In the agents score

• MCP Atlas

  #9 / 38

  MCP Atlas is a benchmark for evaluating AI models on scaled tool use capabilities, measuring how well models can coordinate and utilize multiple tools across complex multi-step tasks.

  79.6%

  Self-reported

• SciCode

  #9 / 30

  SciCode is a research coding benchmark curated by scientists that challenges language models to code solutions for scientific problems. It contains 338 subproblems decomposed from 80 challenging main problems across 16 natural science sub-fields including mathematics, physics, chemistry, biology, and materials science. Problems require knowledge recall, reasoning, and code synthesis skills.

  48.7%

  Self-reported

• AIME 2026

  #5 / 26

  All 30 problems from the 2026 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.

  95.5%

  Self-reported

• ARC-AGI v2

  #11 / 19

  ARC-AGI-2 is an upgraded benchmark for measuring abstract reasoning and problem-solving abilities in AI systems through visual grid transformation tasks. It evaluates fluid intelligence via input-output grid pairs (1x1 to 30x30) using colored cells (0-9), requiring models to identify underlying transformation rules from demonstration examples and apply them to test cases. Designed to be easy for humans but challenging for AI, focusing on core cognitive abilities like spatial reasoning, pattern recognition, and compositional generalization.

  40.1%

  Self-reported · In the text score· In the vision score

• Global-MMLU-Lite

  #5 / 16

  A lightweight version of Global MMLU benchmark that evaluates language models across multiple languages while addressing cultural and linguistic biases in multilingual evaluation.

  86.7%

  Self-reported

• GDPval-AA

  #10 / 14

  GDPval-AA evaluates AI agents on economically valuable professional knowledge-work tasks and reports performance as an Elo score.

  42.3%

  Self-reported

• Artificial Analysis

  #12 / 13

  Artificial Analysis benchmark evaluates AI models across quality, speed, and pricing dimensions, providing a composite assessment of model capabilities for real-world usage.

  40.0%

  Self-reported

• Tau3 Banking

  #9 / 13

  τ³-Bench banking domain evaluates agentic models on multi-turn, tool-using customer-support scenarios in a simulated retail banking environment.

  15.5%

  Self-reported

• ARC-AGI

  #7 / 11

  The Abstraction and Reasoning Corpus for Artificial General Intelligence (ARC-AGI) is a benchmark designed to test general intelligence and abstract reasoning capabilities through visual grid-based transformation tasks. Each task consists of 2-5 demonstration pairs showing input grids transformed into output grids according to underlying rules, with test-takers required to infer these rules and apply them to novel test inputs. The benchmark uses colored grids (up to 30x30) with 10 discrete colors/symbols, designed to measure human-like general fluid intelligence and skill-acquisition efficiency with minimal prior knowledge.

  84.0%

  Self-reported

• CritPT

  #4 / 6

  CritPT is a challenging reasoning benchmark reported by Qwen for evaluating frontier mathematical and critical problem-solving capability.

  8.3%

  Self-reported

• MMAU

  #3 / 5

  A massive multi-task audio understanding and reasoning benchmark comprising 10,000 carefully curated audio clips paired with human-annotated natural language questions spanning speech, environmental sounds, and music. Requires expert-level knowledge and complex reasoning across 27 distinct skills.

  77.0%

  Self-reported

• SimpleQA Verified

  #4 / 5

  SimpleQA Verified is a curated, reliability-focused subset of SimpleQA that addresses label noise and redundancy in the original benchmark, measuring short-form parametric factual accuracy of large language models on fact-seeking questions with single, indisputable answers.

  20.6%

  Self-reported

• AA-Briefcase

  #4 / 4

  AA-Briefcase is an Artificial Analysis evaluation of AI systems on professional knowledge-work tasks, reported as an Elo score.

  30.6%

  Self-reported

• VoiceBench Avg

  #3 / 4

  VoiceBench is the first benchmark designed to provide a multi-faceted evaluation of LLM-based voice assistants, evaluating capabilities including general knowledge, instruction-following, reasoning, and safety using both synthetic and real spoken instruction data with diverse speaker characteristics and environmental conditions.

  90.1%

  Self-reported