Swedish Benchmarks V1.0
Version 1.0
Models answer Swedish language and document questions. The overall score gives task performance a 75% weight and response time a 25% weight by default.
Markovic’s April leaderboard ranks 15 models on ten tests.
- Made by
- Drasko Markovic
- Released
- April 2026
- Kind
- Reading in many languages
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Measures
- Capability
- Subject
- Languages and literature
- Language
- Swedish
- Interaction
- Single turn
- Items from
- Another benchmark
- Refresh
- New versions replace it
- Run by
- A host
- Strongest model tested
-
Qwen3.6 27B #88 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
Gemma 4 31B IT leads with 81.0 among models released by April 2026 Gemma 4 31B IT leads with 81.0 among models released by April 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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 April 2026, 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
- 81.0
Gemma 4 31B IT
- 77.4
Qwen3.6 27B
- 76.1
Qwen 3.6 35B-A3B
- 75.3
Gemma 4 26B A4B
- 74.6 qwen3.5-122b-a10b-halo
- 71.0 gpt-oss-120b
- 70.4
Nemotron 3 Super
- 65.2 gpt-oss-20b
- 63.3 nvidia_nemotron-3-nano
- 61.7
Gemma 3 27B
- 61.2
Llama 3.3 70B
- 60.3
MiniMax-M2.7
- 58.9 glm-4.7-flash
- 45.5 mistralai_ministral-3-14b-reasoning
- 34.1
Mistral Small 4
- 1
Gemma 4 31B IT 81.0
- 2
Qwen3.6 27B 77.4
- 3
Qwen 3.6 35B-A3B 76.1
- 4
Gemma 4 26B A4B 75.3
- 5 qwen3.5-122b-a10b-halo 74.6
- 6 gpt-oss-120b 71.0
- 7
Nemotron 3 Super 70.4
- 8 gpt-oss-20b 65.2
- 9 nvidia_nemotron-3-nano 63.3
- 10
Gemma 3 27B 61.7
- 11
Llama 3.3 70B 61.2
- 12
MiniMax-M2.7 60.3
- 13 glm-4.7-flash 58.9
- 14 mistralai_ministral-3-14b-reasoning 45.5
- 15
Mistral Small 4 34.1
Source: Official Swedish Benchmarks V1.0 leaderboard
- 1
Gemma 4 31B IT 81.0
- 2
Qwen3.6 27B 77.4
- 3
Qwen 3.6 35B-A3B 76.1
- 4
Gemma 4 26B A4B 75.3
- 5 gpt-oss-120b 71.0
- 6
Nemotron 3 Super 70.4
- 7 gpt-oss-20b 65.2
- 8
Gemma 3 27B 61.7
- 9
Llama 3.3 70B 61.2
- 10
MiniMax-M2.7 60.3
- 11 glm-4.7-flash 58.9
- 12
Mistral Small 4 34.1
Source: Official Swedish Benchmarks V1.0 leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 81.0 | Apr 2026 | Official leaderboard | |
| 2 | | 77.4 | Apr 2026 | Official leaderboard | |
| 3 | | 76.1 | Apr 2026 | Official leaderboard | |
| 4 | | 75.3 | Apr 2026 | Official leaderboard | |
| 5 | qwen3.5-122b-a10b-halo | 74.6 | Official leaderboard | ||
| 6 | gpt-oss-120b | 71.0 | Aug 2025 | Official leaderboard | |
| 7 | | 70.4 | Mar 2026 | Official leaderboard | |
| 8 | gpt-oss-20b | 65.2 | Aug 2025 | Official leaderboard | |
| 9 | nvidia_nemotron-3-nano | 63.3 | Official leaderboard | ||
| 10 | | 61.7 | Mar 2025 | Official leaderboard | |
| 11 | | 61.2 | Dec 2024 | Official leaderboard | |
| 12 | | 60.3 | Mar 2026 | Official leaderboard | |
| 13 | glm-4.7-flash | 58.9 | Jan 2026 | Official leaderboard | |
| 14 | mistralai_ministral-3-14b-reasoning | 45.5 | Official leaderboard | ||
| 15 | | 34.1 | Mar 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 81.0 | Apr 2026 | Official leaderboard | |
| 2 | | 77.4 | Apr 2026 | Official leaderboard | |
| 3 | | 76.1 | Apr 2026 | Official leaderboard | |
| 4 | | 75.3 | Apr 2026 | Official leaderboard | |
| 5 | gpt-oss-120b | 71.0 | Aug 2025 | Official leaderboard | |
| 6 | | 70.4 | Mar 2026 | Official leaderboard | |
| 7 | gpt-oss-20b | 65.2 | Aug 2025 | Official leaderboard | |
| 8 | | 61.7 | Mar 2025 | Official leaderboard | |
| 9 | | 61.2 | Dec 2024 | Official leaderboard | |
| 10 | | 60.3 | Mar 2026 | Official leaderboard | |
| 11 | glm-4.7-flash | 58.9 | Jan 2026 | Official leaderboard | |
| 12 | | 34.1 | Mar 2026 | Official leaderboard |
Rankings by part
We read 10 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as New Synonym Benchmark, SweDiagnostics and SweFAQ. Gemma 4 31B IT leads 4 of them, and 5 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
New Synonym Benchmark
12 models we track have a score here.
- 1
Gemma 4 31B IT 98.7
- 2
Gemma 4 26B A4B ↑2 95.1
- 3
Gemma 3 27B ↑5 90.5
- 4
Qwen 3.6 35B-A3B ↓1 89.0
- 5
Qwen3.6 27B ↓3 88.2
- 6
Llama 3.3 70B ↑3 86.3
- 7
Nemotron 3 Super ↓1 85.8
- 8 gpt-oss-20b ↓1 75.3
- 9 gpt-oss-120b ↓4 74.9
- 10
MiniMax-M2.7 63.5
SweDiagnostics
12 models we track have a score here.
- 1
Gemma 4 31B IT 80.7
- 2
Qwen 3.6 35B-A3B ↑1 75.8
- 3 gpt-oss-120b ↑2 73.5
- 4
Nemotron 3 Super ↑2 63.8
- 5
Qwen3.6 27B ↓3 63.5
- 6 gpt-oss-20b ↑1 63.2
- 7
Gemma 3 27B ↑1 61.2
- 8 glm-4.7-flash ↑3 57.9
- 9
Gemma 4 26B A4B ↓5 56.1
- 10
MiniMax-M2.7 47.6
SweFAQ
12 models we track have a score here.
- 1
Qwen3.6 27B ↑1 93.6
- 2 gpt-oss-120b ↑3 92.4
- 3
Qwen 3.6 35B-A3B 92.0
- 4
Nemotron 3 Super ↑2 85.5
- 5
Gemma 4 26B A4B ↓1 78.1
- 6
Gemma 4 31B IT ↓5 77.1
- 7 glm-4.7-flash ↑4 72.4
- 8 gpt-oss-20b ↓1 71.0
- 9
Llama 3.3 70B 70.8
- 10
MiniMax-M2.7 65.4
SweNLI
12 models we track have a score here.
- 1
Gemma 4 31B IT 89.1
- 2 gpt-oss-120b ↑3 83.1
- 3 gpt-oss-20b ↑4 67.2
- 4
Gemma 4 26B A4B 61.1
- 5
Qwen 3.6 35B-A3B ↓2 57.9
- 6
Qwen3.6 27B ↓4 55.6
- 7 glm-4.7-flash ↑4 55.6
- 8
MiniMax-M2.7 ↑2 52.8
- 9
Nemotron 3 Super ↓3 39.7
- 10
Gemma 3 27B ↓2 34.2
SweParaphrase
12 models we track have a score here.
- 1
Gemma 4 26B A4B ↑3 97.7
- 2
Gemma 4 31B IT ↓1 97.6
- 3
Qwen 3.6 35B-A3B 96.8
- 4
Qwen3.6 27B ↓2 95.0
- 5
Gemma 3 27B ↑3 92.5
- 6 glm-4.7-flash ↑5 92.1
- 7
Nemotron 3 Super ↓1 90.5
- 8 gpt-oss-20b ↓1 90.1
- 9
Llama 3.3 70B 89.8
- 10 gpt-oss-120b ↓5 81.7
SwePDF
12 models we track have a score here.
- 1
Qwen 3.6 35B-A3B ↑2 86.8
- 2
Mistral Small 4 ↑10 80.1
- 3
Nemotron 3 Super ↑3 79.1
- 4
Qwen3.6 27B ↓2 77.3
- 5
MiniMax-M2.7 ↑5 75.0
- 6 gpt-oss-120b ↓1 70.9
- 7 gpt-oss-20b 66.7
- 8
Llama 3.3 70B ↑1 44.7
- 9
Gemma 3 27B ↓1 44.5
- 10 glm-4.7-flash ↑1 43.8
SweSAT Synonyms
12 models we track have a score here.
- 1
Gemma 3 27B ↑7 93.3
- 2
Gemma 4 31B IT ↓1 93.0
- 3
Nemotron 3 Super ↑3 93.0
- 4
Llama 3.3 70B ↑5 89.3
- 5
Qwen 3.6 35B-A3B ↓2 89.2
- 6
Gemma 4 26B A4B ↓2 84.0
- 7
Qwen3.6 27B ↓5 74.5
- 8
MiniMax-M2.7 ↑2 72.4
- 9 gpt-oss-120b ↓4 46.6
- 10 glm-4.7-flash ↑1 40.4
SweWiC
12 models we track have a score here.
- 1
Gemma 4 26B A4B ↑3 95.5
- 2
Qwen3.6 27B 92.4
- 3
Gemma 4 31B IT ↓2 92.4
- 4 gpt-oss-120b ↑1 82.2
- 5
MiniMax-M2.7 ↑5 75.0
- 6
Qwen 3.6 35B-A3B ↓3 70.2
- 7
Gemma 3 27B ↑1 61.6
- 8
Nemotron 3 Super ↓2 59.4
- 9 gpt-oss-20b ↓2 56.1
- 10
Llama 3.3 70B ↓1 47.5
SweWinoGender
12 models we track have a score here.
- 1 glm-4.7-flash ↑10 100.0
- 2 gpt-oss-20b ↑5 60.8
- 3
Gemma 4 26B A4B ↑1 53.2
- 4
Gemma 4 31B IT ↓3 48.1
- 5
Mistral Small 4 ↑7 46.0
- 6 gpt-oss-120b ↓1 40.1
- 7
Qwen3.6 27B ↓5 39.4
- 8
Gemma 3 27B 34.6
- 9
Llama 3.3 70B 34.2
- 10
Nemotron 3 Super ↓4 33.1
SweWinograd
12 models we track have a score here.
- 1
Gemma 4 31B IT 97.3
- 2
Qwen3.6 27B 94.8
- 3
Gemma 4 26B A4B ↑1 91.7
- 4
Llama 3.3 70B ↑5 85.2
- 5
Nemotron 3 Super ↑1 74.1
- 6
Qwen 3.6 35B-A3B ↓3 73.0
- 7
MiniMax-M2.7 ↑3 71.5
- 8 gpt-oss-20b ↓1 69.5
- 9 gpt-oss-120b ↓4 64.5
- 10
Gemma 3 27B ↓2 46.0
2 models have held number one on the official leaderboard since its release in April 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Swedish Benchmarks V1.0 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 Swedish Benchmarks V1.0 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.
Versions
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
Data audit
7 of 8 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (15 entries).
- Pass Models scored 12 models we track, plus 3 other entries on its leaderboard.
- Note Newest score The newest came 7 months ago, in April 2026, for Qwen3.6 27B. That’s expected for a test replaced by Swedish Benchmarks V1.1.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Qwen3.6 27B (#88).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on markovic.com: 12 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.