Interactive axes
33
Interactive analysis and open data
Plot any two benchmark boards against each other across shared models with Spearman, Kendall, and Pearson correlations, or download the open redistribution files directly. Build 17 Nov 2026.
33
131
32,061
Hover or focus any mark to inspect exact values on both benchmarks.
Each model as a point in a space of benchmark percentiles, domain composites, principal axes, or domain strengths that remain after the general factor is removed. Choose three axes, drag to rotate, and move the date slider to watch models appear as they were released.
262 models on 33 boards. Missing data rule: a model is plotted only with at least 3 board percentiles. On any axis a model with no value is absent, never filled in.
127 models shown. 135 lack a value on one of these axes. Point size is the number of boards that score the model; faded points have an approximate release date.
X axis, PC1: general factor (66%). Weights: agentic +0.35, coding +0.36, finance +0.32, general-text +0.34, legal +0.26, math +0.31, science +0.39, search +0.3, vision +0.34.
Y axis (up), PC2: +finance -math (20%). Weights: agentic +0.22, coding +0.13, finance +0.49, general-text -0.13, legal +0.49, math -0.46, science -0.08, search -0.44, vision -0.16.
Z axis (depth), PC3: +coding -general text (6%). Weights: agentic +0.48, coding +0.52, finance -0.13, general-text -0.4, legal -0.17, math +0.12, science -0.28, search +0.23, vision -0.39.
| Model | Lab | Released | Weights | Boards | PC1: general factor (66… | PC2: +finance -math (20… | PC3: +coding -general t… |
|---|---|---|---|---|---|---|---|
| GPT-6 Astra | OpenAI | 2026-09-03 | proprietary | 25 | 3.08 | -0.83 | 1.12 |
| Opus 5.5 | Anthropic | 2026-09-22 | proprietary | 23 | 3.86 | 0.7 | 0.25 |
| Gemini 3.8 Flash | 2026-09-02 | proprietary | 23 | 1.48 | -1.78 | -0.84 | |
| Sonnet 5.5 | Anthropic | 2026-09-28 | proprietary | 22 | 3.38 | 0.8 | 0.92 |
| GPT-6.1 Sol | OpenAI | 2026-09-29 | proprietary | 22 | 2.9 | -1.45 | 0.85 |
| GPT-6 Luna | OpenAI | 2026-09-22 | proprietary | 21 | -1.35 | -1.4 | 0.23 |
| Grok 4.7 | xAI | 2026-09-21 | proprietary | 21 | 0.66 | 1.02 | 0.36 |
| Haiku 5.5 | Anthropic | 2026-10-07 | proprietary | 20 | 0.16 | -0.25 | 0.15 |
| GLM-5.3 Flash | Z.ai | 2026-08-20 | open | 20 | 0.47 | -1.75 | -0.85 |
| Gemini 4 Argon | 2026-10-13 | proprietary | 19 | 3.72 | 0.74 | 0.03 | |
| Opus 4.6 | Anthropic | 2026-02-05 | proprietary | 18 | 3.53 | 0.42 | 0.68 |
| Gemini 4 Carbon | 2026-11-10 | proprietary | 18 | 4.3 | 0.75 | 0.46 | |
| Kimi K3 | Moonshot AI | 2026-07-16 | open | 18 | 1.22 | -0.69 | -0.53 |
| GLM-5.3 | Z.ai | 2026-08-14 | open | 18 | 1.8 | -0.18 | -0.5 |
| Opus 4.7 | Anthropic | 2026-04-16 | proprietary | 17 | 3.21 | 0.47 | 0.46 |
| Sonnet 4.6 | Anthropic | 2026-02-17 | proprietary | 17 | 1.38 | -0.09 | 0.23 |
| Gemini 3 Pro Preview | 2025-11-18 | proprietary | 17 | 1.3 | -0.28 | 0.07 | |
| Gemini 3.1 Pro Preview | 2026-02-19 | proprietary | 17 | 2.01 | -0.55 | -0.11 | |
| GPT-5.2 | OpenAI | 2025-12-11 | proprietary | 17 | 0.71 | 0.83 | 0.64 |
| GPT-5.4 | OpenAI | 2026-03-05 | proprietary | 17 | 2.27 | 0.69 | 0.32 |
| Inkling Small | Thinking Machines | 2026-07-30 | open | 17 | -2.69 | -2.17 | -0.81 |
| Fable 5.1 | Anthropic | 2026-09-01 | proprietary | 16 | 3.75 | -0.02 | 0.49 |
| Opus 4.5 | Anthropic | 2025-11-24 | proprietary | 16 | 0.89 | 1.77 | 0.7 |
| Opus 5 | Anthropic | 2026-07-24 | proprietary | 16 | 3.11 | -0.49 | 0.23 |
| Muse Wonder 1.0 | Meta | 2026-10-12 | proprietary | 16 | 2.7 | 0.38 | -0.16 |
| GPT-5.1 | OpenAI | 2025-11-13 | proprietary | 16 | -1.39 | -0.64 | -0.21 |
| GPT-5.6 Sol | OpenAI | 2026-07-09 | proprietary | 16 | 2.86 | -0.98 | 0.55 |
| Inkling | Thinking Machines | 2026-07-15 | open | 16 | -1.73 | -2.63 | -1.14 |
| MiMo-V2.6-Flash | Xiaomi | 2026-09-21 | unknown | 16 | -0.03 | -1.28 | -0.47 |
| MiMo-V2.6-Pro | Xiaomi | 2026-09-21 | unknown | 16 | 1.03 | -0.35 | -0.4 |
| Fable 5 | Anthropic | 2026-06-09 | proprietary | 15 | 3.11 | -0.72 | 0.1 |
| Fable 5.2 | Anthropic | 2026-11-09 | proprietary | 15 | 4.75 | -0.27 | 0.66 |
| Opus 4.8 | Anthropic | 2026-05-28 | proprietary | 15 | 1.6 | 0.19 | -0.16 |
| Sonnet 4.5 | Anthropic | 2025-09-29 | proprietary | 15 | -1.76 | 1.04 | 0.37 |
| Sonnet 5 | Anthropic | 2026-06-30 | proprietary | 15 | 0.55 | 0.66 | 0.04 |
| DeepSeek V4 Pro 0813 | DeepSeek | 2026-08-13 | open | 15 | 0.11 | -1.58 | -0.64 |
| DeepSeek V4.1 Flash 1017 | DeepSeek | 2026-10-17 | open | 15 | 1.92 | -0.15 | 0.13 |
| Gemini 2.5 Pro | 2025-06-05 | proprietary | 15 | -3.1 | -0.23 | -0.49 | |
| Gemini 3 Flash Preview | 2025-12-17 | proprietary | 15 | 0.37 | 1.03 | 0.43 | |
| Gemini 4 Hydrogen | 2026-11-12 | proprietary | 15 | 3.47 | 0.6 | 0.38 | |
| Kimi K2.5 | Moonshot AI | 2026-01-27 | open | 15 | -0.3 | 0.23 | 0.18 |
| Kimi K3.1 Pro | Moonshot AI | 2026-10-23 | open | 15 | 2.35 | -1.35 | 0.13 |
| Step 5 Preview | StepFun | 2026-09-18 | unknown | 15 | -0.22 | -1.14 | -0.58 |
| GPT-5 | OpenAI | 2025-08-07 | proprietary | 14 | -0.16 | 1.09 | 0.34 |
| GPT-5.5 | OpenAI | 2026-04-23 | proprietary | 14 | 2.3 | -0.85 | 0.29 |
| GPT-5.6 Terra | OpenAI | 2026-07-09 | proprietary | 14 | 1.01 | -1.13 | -0.2 |
| GPT-6.1 Terra | OpenAI | 2026-10-26 | proprietary | 14 | 5.48 | -4.28 | -0.51 |
| Beam 1.0 | Reflection | 2026-10-16 | open | 14 | -1.22 | -2.52 | -1.01 |
| Inkling 2 Small | Thinking Machines | 2026-11-05 | unknown | 14 | -0.36 | 1.1 | 0.07 |
| Grok 4.5 | xAI | 2026-07-08 | proprietary | 14 | 0.31 | -0.83 | 0.22 |
| Grok 4.6 | xAI | 2026-08-12 | proprietary | 14 | 0.82 | -0.18 | 0.11 |
| GPT-5.6 Luna | OpenAI | 2026-07-09 | proprietary | 13 | -0.17 | -0.07 | 0.07 |
| GPT-6 Sol | OpenAI | 2026-09-22 | proprietary | 13 | 2.43 | 0.91 | 0.71 |
| o3 | OpenAI | 2025-04-16 | proprietary | 13 | -1.77 | -0.28 | -0.24 |
| Opus 4 | Anthropic | 2025-05-22 | proprietary | 12 | -2.26 | 1.6 | 0.12 |
| Gemini 3.5 Flash | 2026-05-19 | proprietary | 12 | 2.05 | 0.42 | 0.42 | |
| Gemini 3.7 Flash | 2026-08-13 | proprietary | 12 | 1.52 | -2.02 | -0.39 | |
| Muse Spark Open | Meta | 2026-10-15 | open | 12 | -1.6 | -1.16 | -0.7 |
| Muse Wonder Full 1.0 | Meta | 2026-10-12 | unknown | 12 | 1.94 | 1.67 | 0.78 |
| Hy4 Preview | Tencent | 2026-08-28 (approx.) | unknown | 12 | -0.81 | -0.76 | -0.44 |
| Claude 3.7 Sonnet | Anthropic | 2025-02-24 | proprietary | 11 | -2.94 | 2.2 | 0.28 |
| Haiku 4.5 | Anthropic | 2025-10-15 | proprietary | 11 | -4.05 | 1.26 | 0.04 |
| Opus 4.1 | Anthropic | 2025-08-05 | proprietary | 11 | -1.09 | 1.19 | -0.26 |
| Muse Ember | Meta | 2026-11-09 | unknown | 11 | 0.12 | 1.72 | 0.57 |
| MiniMax M3 | MiniMax | 2026-06-01 | open | 11 | -2.44 | -1.1 | -0.58 |
| MiniMax M3.1 Flash | MiniMax | 2026-10-19 | unknown | 11 | -0.38 | -1.41 | -0.41 |
| Kimi K2.6 | Moonshot AI | 2026-04-20 | open | 11 | 0.99 | 0.88 | 0.41 |
| GPT-5 mini | OpenAI | 2025-08-07 | proprietary | 11 | -2.26 | -0.16 | -0.1 |
| Qwen3.8-Max | Alibaba | 2026-08-02 | proprietary | 10 | 1.92 | -2.58 | -0.49 |
| Gemini 3.6 Flash | 2026-07-21 | proprietary | 10 | 0.44 | -0.72 | -0.07 | |
| Muse Spark 1.3 | Meta | 2026-09-02 | unknown | 10 | 2.08 | -0.79 | 0.04 |
| Mistral Large 4 OSS | Mistral | 2026-10-29 | open | 10 | 0 | -0.67 | -0.13 |
| Hy4 | Tencent | 2026-11-03 | unknown | 10 | 1.46 | -0.37 | 0.13 |
| Qwen3.7 Plus | Alibaba | 2026-06-01 | proprietary | 9 | -1.83 | -2.19 | -0.83 |
| Qwen3.8 Max Internal | Alibaba | 2026-11-15 (approx.) | unknown | 9 | 0.41 | 0.65 | 0.26 |
| Qwen4 Max | Alibaba | 2026-11-15 (approx.) | proprietary | 9 | 1.49 | 0.88 | 0.47 |
| Claude Sonnet 4 | Anthropic | 2025-05-26 (approx.) | proprietary | 9 | -1.95 | 2.6 | 0.41 |
| DeepSeek V4.1 Flash | DeepSeek | 2026-09-09 | open | 9 | 1.45 | -0.56 | -0.02 |
| Gemma 4 31B | 2026-04-02 | open | 9 | -1.48 | -2.16 | -0.77 | |
| Mercury 3 | Inception | 2026-11-15 (approx.) | unknown | 9 | -1.82 | -2.02 | -0.8 |
| Muse Spark | Meta | 2026-04-08 | proprietary | 9 | 1.37 | -1.31 | -0.35 |
| Muse Spark 1.1 | Meta | 2026-07-09 | proprietary | 9 | 0.95 | -1.84 | -0.46 |
| Mistral Large 4 | Mistral | 2026-10-06 | open | 9 | -1.2 | -0.72 | -0.34 |
| Nemotron 3.5 Super VL | NVIDIA | 2026-10-15 | open | 9 | -2.29 | -2.3 | -0.96 |
| GPT-4.1 | OpenAI | 2025-04-14 | proprietary | 9 | -2.69 | 2.68 | 0.42 |
| GPT-5.4 mini | OpenAI | 2026-03-17 | proprietary | 9 | -1.13 | -0.71 | -0.37 |
| Grok 4.3 | xAI | 2026-04-30 | proprietary | 9 | -0.27 | 0.73 | 0.14 |
| GLM-5 | Z.ai | 2026-02-11 | open | 9 | -0.19 | -0.32 | -0.25 |
| GLM-5.2 | Z.ai | 2026-06-16 | open | 9 | 1.4 | -0.12 | 0.03 |
| Qwen3.8-27B | Alibaba | 2026-08-14 | open | 8 | -0.74 | -1.79 | -0.61 |
| Claude 3.5 Sonnet (Oct '24) | Anthropic | 2024-10-22 | proprietary | 8 | -2.35 | 3.81 | 0.76 |
| DeepSeek V4-Pro | DeepSeek | 2026-04-24 | open | 8 | 0.48 | -2.87 | -0.8 |
| GPT-5 nano | OpenAI | 2025-08-07 | proprietary | 8 | -2.3 | 2.57 | 0.51 |
| o1 | OpenAI | 2024-12-05 | proprietary | 8 | -1.36 | 2.52 | 0.55 |
| o4-mini | OpenAI | 2025-04-16 | proprietary | 8 | -1.92 | 0.56 | 0.12 |
| Grok 4 (0709) | xAI | 2025-07-09 | proprietary | 8 | -0.67 | 1.55 | 0.39 |
| Grok 4.1 Fast | xAI | 2025-11-19 | proprietary | 8 | -0.29 | 1 | 0.53 |
| Qwen3.6 Max Preview | Alibaba | 2026-04-20 | unknown | 7 | 1.62 | -0.76 | -0.04 |
| Qwen3.6 Plus | Alibaba | 2026-04-02 | proprietary | 7 | -0.09 | -2.02 | -0.64 |
| DeepSeek V3.2 | DeepSeek | 2025-12-01 | open | 7 | -1.12 | 0.22 | -0.15 |
| Gemini 1.5 Pro (002) | 2024-09-24 | proprietary | 7 | -3.5 | 2.88 | 0.45 | |
| Gemini 2.0 Flash (001) | 2025-02-05 | proprietary | 7 | -1.86 | 2.67 | 0.57 | |
| Gemini 2.5 Flash | 2025-05-20 | proprietary | 7 | -2.13 | -1.45 | -0.7 | |
| Gemini 3.1 Flash-Lite Preview | 2026-03-03 | proprietary | 7 | -0.27 | 1.2 | 0.29 | |
| Gemini 3.5 Flash-Lite | 2026-07-21 | proprietary | 7 | 0.51 | 0.36 | 0.11 | |
| Llama 4 Scout | Meta | 2025-04-05 | open | 7 | -3.13 | 3.31 | 0.63 |
| Muse Spark 1.2 | Meta | 2026-08-05 | unknown | 7 | 0.68 | -4.01 | -1.15 |
| Mistral Medium 3 | Mistral | 2025-05-07 | proprietary | 7 | -1.02 | 4.64 | 1.01 |
| GPT-4.1 mini | OpenAI | 2025-04-14 | proprietary | 7 | -1.95 | 2.49 | 0.49 |
| GPT-4.1 nano | OpenAI | 2025-04-14 | proprietary | 7 | -3.09 | 3.77 | 0.77 |
| GPT-4.5 Preview | OpenAI | 2025-02-28 (approx.) | proprietary | 7 | -2.41 | 2.38 | 0.39 |
| GPT-5.4 nano | OpenAI | 2026-03-17 | proprietary | 7 | -1.29 | 1.81 | 0.39 |
| GLM-4.7 | Z.ai | 2025-12-22 | open | 7 | 0.05 | -1.76 | -0.52 |
| GLM-5.1 | Z.ai | 2026-04-07 | open | 7 | 1.61 | -1.25 | -0.21 |
| Llama 4 Maverick | Meta | 2025-04-05 | open | 6 | -3.63 | 3.28 | 0.55 |
| MiniMax M2.5 | MiniMax | 2026-02-12 | open | 6 | -1.85 | 0.4 | -0.03 |
| Mistral Medium 3.5 | Mistral | 2026-04-29 | unknown | 6 | -2.2 | -1.47 | -0.75 |
| Nemotron 3.5 Super VL 120B A12B (NVFP4) | NVIDIA | 2026-11-09 (approx.) | open | 6 | -1.95 | -0.71 | -0.47 |
| GPT-4o mini | OpenAI | 2024-07-18 | proprietary | 6 | -3.16 | 3.95 | 0.8 |
| gpt-oss-120b | OpenAI | 2025-08-05 | open | 6 | -3.1 | -1.2 | -0.7 |
| Grok 4.20 Multi-Agent Beta (0309) | xAI | 2026-11-09 (approx.) | proprietary | 6 | 1.26 | -0.06 | 0.14 |
| MiniMax M2 | MiniMax | 2025-10-26 | open | 5 | -2.74 | 0.21 | -0.23 |
| GPT-4o (Nov '24) | OpenAI | 2024-11-20 | proprietary | 5 | -5.68 | 3.58 | 0.58 |
| gpt-oss-20b | OpenAI | 2025-08-05 | open | 5 | -4.32 | -0.05 | -0.49 |
| DeepSeek V4-Flash 0731 | DeepSeek | 2026-07-31 | open | 4 | -1.23 | -2.58 | -0.89 |
| Gemini 2.5 Pro Preview (06-05) | 2025-06-05 | proprietary | 4 | -3.01 | -0.04 | -0.2 | |
| GPT-5.4 Pro | OpenAI | 2026-03-05 | proprietary | 4 | 1.1 | -1.66 | -0.37 |
Method rules
Known limits
Exclusions
Static CSV, JSONL, and search-index artifacts emitted during build. Restricted sources (Artificial Analysis) are automatically excluded from bulk downloads.
Flat relational table of every non-restricted result row and interval.
Download acropolis-open-results.csvLine-delimited JSON objects with source provenance and cost fields.
Download acropolis-open-results.jsonlCanonical catalog of evaluated models, benchmarks, and study links.
Download search-index.jsonDuckDB quick query
SELECT benchmarkId, modelId, value, ciLow, ciHigh
FROM read_csv_auto('acropolis-open-results.csv')
WHERE benchmarkId = 'swe-bench-verified'
ORDER BY value DESC
LIMIT 15;Python Polars quick query
import polars as pl
df = pl.read_ndjson("acropolis-open-results.jsonl")
print(df.group_by("benchmarkId").len().sort("len", descending=True))