Code2Video-AA
As of 29 Oct 2026, the top score on Code2Video-AA is 1536 by Claude Opus 5.5 (Anthropic), ahead of Claude Sonnet 5.5 at 1443, across 27 model configurations with a published result (Artificial Analysis, result dated 22 Sept 2026). Artificial Analysis' run of HeyGen's Code2Video Bench: short product and brand videos built in HTML from 114 design briefs, graded head to head, so the score is an Elo. Code2Video-AA is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 29 Oct 2026)
- Claude Opus 5.5 holds the top Code2Video-AA score at 1536 (Artificial Analysis, 22 Sept 2026).
- Claude Sonnet 5.5 is second on Code2Video-AA at 1443 (Artificial Analysis, 28 Sept 2026).
- Code2Video-AA: Artificial Analysis' run of HeyGen's Code2Video Bench: short product and brand videos built in HTML from 114 design briefs, graded head to head, so the score is an Elo (Artificial Analysis).
- 27 model configurations have a published Code2Video-AA result on BenchLeader (BenchLeader, 29 Oct 2026).
- Code2Video-AA does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Artificial Analysis' run of HeyGen's Code2Video Bench: short product and brand videos built in HTML from 114 design briefs, graded head to head, so the score is an Elo.
- Published by
- Artificial Analysis
- Category
- Multimodal
- Index weight
- Reference only
- Models
- 27
- Data as of
- 29 Oct 2026
- Newest result
- 16 Oct 2026
Source: Artificial Analysis (artificialanalysis.ai). Data taken from the public leaderboard.
- 1Claude Opus 5.5 (max)1536
- 2Claude Sonnet 5.5 (max)1443
- 3GPT-6 Astra (max)1427
- 4GPT-6.1 Sol (max)1369
- 5Claude Fable 5.1 (max)1325
- 6Gemini 4 Argon (high)1259
- 7Claude Haiku 5.5 (max)1192
- 8Grok 4.7 (xhigh)1127
- 9DeepSeek V4.1 Flash (max)1053
- 10MiMo-V2.6-Flash1052
- 11MiMo-V2.6-Pro1052
- 12Kimi K3 (max)1041
- 13GLM 5.3 (max)1034
- 14GPT-6 Luna (max)1030
- 15Qwen3.8 Max (max)1021
| # | ||||
|---|---|---|---|---|
| 1 |  Claude Opus 5.5max | 1536 | 69.2 | 2026-09-22 |
| 2 | Claude Sonnet 5.5max | 1443 | 65.0 | 2026-09-28 |
| 3 |  GPT-6 Astramax | 1427 | 67.7 | 2026-09-03 |
| 4 | GPT-6.1 Solmax | 1369 | 67.2 | 2026-09-29 |
| 5 | Claude Fable 5.1max | 1325 | 67.0 | 2026-09-01 |
| 6 |  Gemini 4 Argonhigh | 1259 | 65.9 | 2026-09-30 |
| 7 | Claude Haiku 5.5max | 1192 | 56.7 | 2026-10-07 |
| 8 |  Grok 4.7xhigh | 1127 | 58.1 | 2026-09-21 |
| 9 |  DeepSeek V4.1 Flashmaxopen ↗ | 1053 | 58.9 | 2026-09-10 |
| 10 |  MiMo-V2.6-Flashopen ↗ | 1052 | 55.0 | 2026-09-21 |
| 11 | MiMo-V2.6-Proopen ↗ | 1052 | 57.2 | 2026-09-21 |
| 12 |  Kimi K3maxopen ↗ | 1041 | 62.2 | 2026-07-16 |
| 13 |  GLM 5.3maxopen ↗ | 1034 | 58.9 | 2026-08-18 |
| 14 | GPT-6 Lunamax | 1030 | 56.6 | 2026-09-22 |
| 15 |  Qwen3.8 Maxmax | 1021 | 58.8 | 2026-09-02 |
| 16 |  Muse Spark 1.3max | 1019 | 61.8 | 2026-09-02 |
| 17 | GPT-5.5xhigh | 1000 | 64.4 | 2026-04-23 |
| 18 | Muse Wondermax | 997 | 62.2 | 2026-10-12 |
| 19 | GLM 5.3 Flashopen ↗ | 975 | 57.2 | 2026-08-26 |
| 20 | Qwen3.8 27Bxhighopen ↗ | 949 | 52.2 | 2026-08-14 |
| 21 |  Mistral Large 4open | 931 | 50.5 | 2026-10-06 |
| 22 | Gemini 3.8 Flashhigh | 899 | 61.9 | 2026-09-02 |
| 23 |  Step 5 Previewopen | 821 | 54.8 | 2026-09-18 |
| 24 |  MiniMax M3open ↗ | 724 | 56.3 | 2026-06-01 |
| 25 |  Beam 1.0open | 675 | 45.2 | 2026-10-16 |
| 26 | Inklingxhighopen ↗ | 467 | 49.8 | 2026-07-15 |
| 27 |  Nemotron 3 Ultra 550B A55Bthinkingopen ↗ | 304 | 48.5 | 2026-06-04 |
More multimodal benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on Code2Video-AA
Cite as: BenchLeader, “Code2Video-AA leaderboard”, data as of 29 Oct 2026, https://www.benchleader.com/benchmarks/aa_code2video.
Code2Video-AA: questions
- What does Code2Video-AA measure?
- Artificial Analysis' run of HeyGen's Code2Video Bench: short product and brand videos built in HTML from 114 design briefs, graded head to head, so the score is an Elo. Scores are reported in Elo-style ratings from pairwise comparisons; higher is better.
- What is the top score on Code2Video-AA?
- As of 29 Oct 2026, the highest score on Code2Video-AA is 1536 by Claude Opus 5.5 (result dated 22 Sept 2026), ahead of Claude Sonnet 5.5 at 1443. Scores come from Artificial Analysis and are re-read daily.
- Which AI model leads Code2Video-AA?
- Claude Opus 5.5 leads Code2Video-AA with 1536 as of 29 Oct 2026, ahead of Claude Sonnet 5.5 at 1443 (result dated 22 Sept 2026).
- How many models have Code2Video-AA results?
- 27 model configurations have a Code2Video-AA result on BenchLeader, all taken from Artificial Analysis.
- Who runs Code2Video-AA and how often is it updated?
- Code2Video-AA is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Code2Video-AA count toward the BenchLeader Index?
- No. Code2Video-AA is shown for reference but left out of the composite index.