Swallow - Japanese MT-Bench - STEM: leaderboard
From the Swallow LLM Leaderboard: Japanese MT-Bench (Nejumi edition, v4 questions with corrected v2 reference answers), two-turn dialogues graded by GPT-5.2 against references; normalized 0-1 upstream, shown as 0-100, higher is better. This board covers the STEM category: explanations in physics, chemistry, biology, geography, architecture and machine learning.
Metric: Judge Score (normalized, %). Source: swallow-llm.github.io. Saturation forecast: Estimated already saturated. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Mini (2025-08-07) (Medium) | 83.8 | #175 (GPT-5 Mini (2025-08-07)) |
| 2 | GPT-5.4 (2026-03-05) (Medium) | 83.2 | #63 (GPT-5.4 (2026-03-05)) |
| 3 | GPT-5 (Medium) | 83.1 | #105 (GPT-5) |
| 4 | O3 (2025-04-16) (Medium) | 80.5 | #125 (O3 (2025-04-16)) |
| 5 | Gemma 4 31B (IT) (Thinking) | 77.9 | #178 (Gemma 4 31B (IT)) |
| 6 | Qwen 3.5 122B A10B (Thinking) | 77.1 | #184 (Qwen 3.5 122B A10B) |
| 7 | Qwen 3.5 27B (Thinking) | 76.5 | #192 (Qwen 3.5 27B) |
| 8 | Qwen 3.5 35B A3B (Thinking) | 76 | #268 (Qwen 3.5 35B A3B) |
| 9 | gemma-4-26B-A4B-it (Thinking) | 75.9 | |
| 10 | Qwen 3 235B A22B 2507 Instruct | 75.6 | #315 |
| 11 | O3 Mini (Medium) | 73.7 | #282 (O3 Mini) |
| 12 | Qwen 3 Next 80B A3B (Thinking) | 73 | #312 |
| 13 | GPT-OSS-Swallow-120B-RL-v0.1 (Thinking) | 72.9 | #309 (GPT-OSS-Swallow-120B-RL-v0.1) |
| 14 | Qwen 3 235B A22B 2507 (Thinking) | 72.8 | #263 |
| 15 | Qwen 3 Next 80B A3B Instruct | 72.4 | #381 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=swallow-japanese-mt-bench-stem · How It Works · Data refreshed daily, snapshot 2026-11-05.