InteractScience
Models generate code for interactive scientific demonstrations. Evaluators test the program and assess its visual output.
- Made by
- OpenCompass and collaborators
- Released
- October 2025
- Kind
- Scientific code
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-5 #71 on the capability index
- Last model added
- September 2025
- Scores refreshed
- November 20th, 2026
Claude Sonnet 4 leads with 41.5 among models released by September 2025 Claude Sonnet 4 leads with 41.5 among models released by September 2025
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The authors report separate functional and visual measures. The headline is the overall functional test pass rate. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 41.5 Claude Sonnet 4
- 40.3 Claude Opus 4
- 39.5 GPT-5
- 37.3 o4-mini
- 37.1 GPT-4.1
- 35.3 Gemini-2.5-Pro (snapshot unspecified)
- 34.9 o3
- 33.9
DeepSeek-R1
- 33.3 Claude-3.5-Sonnet (snapshot unspecified)
- 33.3
Qwen3-235B-A22B-Instruct
- 31.9
Intern-S1
- 31.7
DeepSeek-V3
- 31.6 Kimi K2
- 31.6
Gemini 2.5 Flash
- 29.3 GLM-4.5
- 28.3 GPT-4o (snapshot unspecified)
- 28.0 gpt-oss-120b
- 27.2
Qwen2.5-Coder-32B-Instruct
- 27.2
Qwen3-32B
- 24.1
Qwen3-14B
- 23.7
Qwen2.5-VL-72B
- 22.5 Qwen2.5-Coder-14B-Instruct
- 20.0
Qwen3-8B
- 18.7
Llama 3.1-70B
- 15.2 gpt-oss-20b
- 14.7
Qwen3-4B
- 12.4
Qwen2.5-Coder (7B)
- 11.3
Llama 3.1-8B
- 7.47
Qwen2.5-VL-7B
- 6.53
Qwen3-1.7B
- 1 Claude Sonnet 4 41.5
- 2 Claude Opus 4 40.3
- 3 GPT-5 39.5
- 4 o4-mini 37.3
- 5 GPT-4.1 37.1
- 6 Gemini-2.5-Pro (snapshot unspecified) 35.3
- 7 o3 34.9
- 8
DeepSeek-R1 May 2025 33.9
- 9 Claude-3.5-Sonnet (snapshot unspecified) 33.3
- 10
Qwen3-235B-A22B-Instruct Jul 2025 33.3
- 11
Intern-S1 31.9
- 12
DeepSeek-V3 Mar 2025 31.7
- 13 Kimi K2 Sep 2025 31.6
- 14
Gemini 2.5 Flash Sep 2025 31.6
- 15 GLM-4.5 29.3
- 16 GPT-4o (snapshot unspecified) 28.3
- 17 gpt-oss-120b 28.0
- 18
Qwen2.5-Coder-32B-Instruct 27.2
- 19
Qwen3-32B 27.2
- 20
Qwen3-14B 24.1
- 21
Qwen2.5-VL-72B 23.7
- 22 Qwen2.5-Coder-14B-Instruct 22.5
- 23
Qwen3-8B 20.0
- 24
Llama 3.1-70B 18.7
- 25 gpt-oss-20b 15.2
- 26
Qwen3-4B 14.7
- 27
Qwen2.5-Coder (7B) 12.4
- 28
Llama 3.1-8B 11.3
- 29
Qwen2.5-VL-7B 7.47
- 30
Qwen3-1.7B 6.53
Source: Official InteractScience leaderboard
- 41.5 Claude Sonnet 4
- 40.3 Claude Opus 4
- 39.5 GPT-5
- 37.3 o4-mini
- 37.1 GPT-4.1
- 34.9 o3
- 33.9
DeepSeek-R1
- 33.3
Qwen3-235B-A22B-Instruct
- 31.9
Intern-S1
- 31.7
DeepSeek-V3
- 31.6 Kimi K2
- 31.6
Gemini 2.5 Flash
- 29.3 GLM-4.5
- 28.0 gpt-oss-120b
- 27.2
Qwen2.5-Coder-32B-Instruct
- 27.2
Qwen3-32B
- 24.1
Qwen3-14B
- 23.7
Qwen2.5-VL-72B
- 22.5 Qwen2.5-Coder-14B-Instruct
- 20.0
Qwen3-8B
- 18.7
Llama 3.1-70B
- 15.2 gpt-oss-20b
- 14.7
Qwen3-4B
- 12.4
Qwen2.5-Coder (7B)
- 11.3
Llama 3.1-8B
- 7.47
Qwen2.5-VL-7B
- 6.53
Qwen3-1.7B
- 1 Claude Sonnet 4 41.5
- 2 Claude Opus 4 40.3
- 3 GPT-5 39.5
- 4 o4-mini 37.3
- 5 GPT-4.1 37.1
- 6 o3 34.9
- 7
DeepSeek-R1 May 2025 33.9
- 8
Qwen3-235B-A22B-Instruct Jul 2025 33.3
- 9
Intern-S1 31.9
- 10
DeepSeek-V3 Mar 2025 31.7
- 11 Kimi K2 Sep 2025 31.6
- 12
Gemini 2.5 Flash Sep 2025 31.6
- 13 GLM-4.5 29.3
- 14 gpt-oss-120b 28.0
- 15
Qwen2.5-Coder-32B-Instruct 27.2
- 16
Qwen3-32B 27.2
- 17
Qwen3-14B 24.1
- 18
Qwen2.5-VL-72B 23.7
- 19 Qwen2.5-Coder-14B-Instruct 22.5
- 20
Qwen3-8B 20.0
- 21
Llama 3.1-70B 18.7
- 22 gpt-oss-20b 15.2
- 23
Qwen3-4B 14.7
- 24
Qwen2.5-Coder (7B) 12.4
- 25
Llama 3.1-8B 11.3
- 26
Qwen2.5-VL-7B 7.47
- 27
Qwen3-1.7B 6.53
Source: Official InteractScience leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 41.5 | May 2025 | Official leaderboard | |
| 2 | Claude Opus 4 | 40.3 | May 2025 | Official leaderboard | |
| 3 | GPT-5 | 39.5 | Aug 2025 | Official leaderboard | |
| 4 | o4-mini | 37.3 | Apr 2025 | Official leaderboard | |
| 5 | GPT-4.1 | 37.1 | Apr 2025 | Official leaderboard | |
| 6 | Gemini-2.5-Pro (snapshot unspecified) | 35.3 | Official leaderboard | ||
| 7 | o3 | 34.9 | Apr 2025 | Official leaderboard | |
| 8 | | 33.9 | May 2025 | Official leaderboard | |
| 9 | Claude-3.5-Sonnet (snapshot unspecified) | 33.3 | Official leaderboard | ||
| 10 | | 33.3 | Jul 2025 | Official leaderboard | |
| 11 | | 31.9 | Jul 2025 | Official leaderboard | |
| 12 | | 31.7 | Mar 2025 | Official leaderboard | |
| 13 | Kimi K2 Sep 2025 | 31.6 | Sep 2025 | Official leaderboard | |
| 14 | | 31.6 | Sep 2025 | Official leaderboard | |
| 15 | GLM-4.5 | 29.3 | Aug 2025 | Official leaderboard | |
| 16 | GPT-4o (snapshot unspecified) | 28.3 | Official leaderboard | ||
| 17 | gpt-oss-120b | 28.0 | Aug 2025 | Official leaderboard | |
| 18 | | 27.2 | Nov 2024 | Official leaderboard | |
| 19 | | 27.2 | Apr 2025 | Official leaderboard | |
| 20 | | 24.1 | Apr 2025 | Official leaderboard | |
| 21 | | 23.7 | Jan 2025 | Official leaderboard | |
| 22 | Qwen2.5-Coder-14B-Instruct | 22.5 | Nov 2024 | Official leaderboard | |
| 23 | | 20.0 | Apr 2025 | Official leaderboard | |
| 24 | | 18.7 | Jul 2024 | Official leaderboard | |
| 25 | gpt-oss-20b | 15.2 | Aug 2025 | Official leaderboard | |
| 26 | | 14.7 | Apr 2025 | Official leaderboard | |
| 27 | | 12.4 | Sep 2024 | Official leaderboard | |
| 28 | | 11.3 | Jul 2024 | Official leaderboard | |
| 29 | | 7.47 | Jan 2025 | Official leaderboard | |
| 30 | | 6.53 | Apr 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 41.5 | May 2025 | Official leaderboard | |
| 2 | Claude Opus 4 | 40.3 | May 2025 | Official leaderboard | |
| 3 | GPT-5 | 39.5 | Aug 2025 | Official leaderboard | |
| 4 | o4-mini | 37.3 | Apr 2025 | Official leaderboard | |
| 5 | GPT-4.1 | 37.1 | Apr 2025 | Official leaderboard | |
| 6 | o3 | 34.9 | Apr 2025 | Official leaderboard | |
| 7 | | 33.9 | May 2025 | Official leaderboard | |
| 8 | | 33.3 | Jul 2025 | Official leaderboard | |
| 9 | | 31.9 | Jul 2025 | Official leaderboard | |
| 10 | | 31.7 | Mar 2025 | Official leaderboard | |
| 11 | Kimi K2 Sep 2025 | 31.6 | Sep 2025 | Official leaderboard | |
| 12 | | 31.6 | Sep 2025 | Official leaderboard | |
| 13 | GLM-4.5 | 29.3 | Aug 2025 | Official leaderboard | |
| 14 | gpt-oss-120b | 28.0 | Aug 2025 | Official leaderboard | |
| 15 | | 27.2 | Nov 2024 | Official leaderboard | |
| 16 | | 27.2 | Apr 2025 | Official leaderboard | |
| 17 | | 24.1 | Apr 2025 | Official leaderboard | |
| 18 | | 23.7 | Jan 2025 | Official leaderboard | |
| 19 | Qwen2.5-Coder-14B-Instruct | 22.5 | Nov 2024 | Official leaderboard | |
| 20 | | 20.0 | Apr 2025 | Official leaderboard | |
| 21 | | 18.7 | Jul 2024 | Official leaderboard | |
| 22 | gpt-oss-20b | 15.2 | Aug 2025 | Official leaderboard | |
| 23 | | 14.7 | Apr 2025 | Official leaderboard | |
| 24 | | 12.4 | Sep 2024 | Official leaderboard | |
| 25 | | 11.3 | Jul 2024 | Official leaderboard | |
| 26 | | 7.47 | Jan 2025 | Official leaderboard | |
| 27 | | 6.53 | Apr 2025 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as PFT Overall, VQT Action and VQT CLIP. Claude Sonnet 4 leads 2 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
PFT Overall
27 models we track have a score here.
- 1 Claude Sonnet 4 41.5%
- 2 Claude Opus 4 40.3%
- 3 GPT-5 39.5%
- 4 o4-mini 37.3%
- 5 GPT-4.1 37.1%
- 6 o3 34.9%
- 7
DeepSeek-R1 May 2025 33.9%
- 8
Qwen3-235B-A22B-Instruct Jul 2025 33.3%
- 9
Intern-S1 31.9%
- 10
DeepSeek-V3 Mar 2025 31.7%
VQT Action
27 models we track have a score here.
- 1 gpt-oss-120b ↑13 90.8%
- 2 o3 ↑4 89.8%
- 3 Claude Sonnet 4 ↓2 89.7%
- 4 GPT-5 ↓1 89.7%
- 5 Claude Opus 4 ↓3 89.3%
- 6 GPT-4.1 ↓1 89.2%
- 7
Llama 3.1-70B ↑14 88.6%
- 8 o4-mini ↓4 88.6%
- 9
DeepSeek-R1 May 2025 ↓2 88.3%
- 10
Qwen3-32B ↑6 87.5%
VQT CLIP
27 models we track have a score here.
- 1 Claude Sonnet 4 73.5
- 2 Claude Opus 4 73.2
- 3 o3 ↑3 72.2
- 4 gpt-oss-120b ↑10 72.1
- 5 GPT-5 ↓2 72.0
- 6 o4-mini ↓2 71.8
- 7 GPT-4.1 ↓2 71.2
- 8 Kimi K2 Sep 2025 ↑3 70.1
- 9
Qwen3-235B-A22B-Instruct Jul 2025 ↓1 70.0
- 10
Gemini 2.5 Flash Sep 2025 ↑2 69.6
VQT VLM-judge
27 models we track have a score here.
- 1 GPT-5 ↑2 57.0
- 2 Claude Sonnet 4 ↓1 55.4
- 3 Claude Opus 4 ↓1 54.9
- 4 GPT-4.1 ↑1 52.8
- 5 o3 ↑1 52.8
- 6 o4-mini ↓2 51.9
- 7 Kimi K2 Sep 2025 ↑4 50.0
- 8 gpt-oss-120b ↑6 49.6
- 9
DeepSeek-V3 Mar 2025 ↑1 49.5
- 10
DeepSeek-R1 May 2025 ↓3 49.5
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official InteractScience leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official InteractScience leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (30 entries).
- Pass Models scored 27 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 27 rows for models we track and 3 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.