FlavourBench Powered v43
Version powered v43
Models choose ingredients for 640 tasks across four categories, including regional composition. Failed responses earn zero. Models must complete 608 tasks to qualify for ranking.
The archived table ranks 25 of 26 models. Claude Fable 5 completed 242 tasks and received no rank.
- Made by
- Josef Chen, Jakub Radzikowski, and Erim Hayretci
- Released
- August 2026
- Kind
- Taste and style
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Generated
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- FlavourBench Complete Core
- Strongest model tested
- Claude Opus 5 #11 on the capability index
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
Grok 4.6 leads with 73.4 among models released by August 2026 Grok 4.6 leads with 73.4 among models released by August 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 August 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 15 more
- 73.4 Grok 4.6
- 72.9 Google: Gemini 3.1 Pro Preview
- 72.4
Muse Spark 1.2
- 72.0 GPT-5.6 Sol Pro
- 71.5 OpenAI: GPT-5.6 Terra Pro
- 71.4
Qwen 3.8 Max
- 71.2
Qwen3.8-2.4T-A95B
- 70.8 Claude Opus 5
- 70.3 Kimi K3
- 70.1
Gemini 3.6 Flash
- 69.4 Claude Sonnet 5
- 69.2 ByteDance Seed: Seed 2.1 Turbo
- 69.2 OpenAI: GPT-5.6 Luna Pro
- 68.9
Tencent Hy3
- 68.5
DeepSeek V4 Pro 0813
- 68.4 Meta: Muse Glimmer 30B
- 66.9 GLM-5.2
- 66.7
MiniMax-M3
- 65.2 NVIDIA: Nemotron 3.5 Lightning
- 64.5
Inkling
- 62.7
DeepSeek V4 Flash 0731
- 61.4 Mistral: Mistral Large 3 2512
- 60.2
Llama 4 Maverick
- 58.1
Command A+
- 36.7 Cohere: Command R+ (08-2024)
- 1 Grok 4.6 73.4
- 2 Google: Gemini 3.1 Pro Preview 72.9
- 3
Muse Spark 1.2 72.4
- 4 GPT-5.6 Sol Pro 72.0
- 5 OpenAI: GPT-5.6 Terra Pro 71.5
- 6
Qwen 3.8 Max 71.4
- 7
Qwen3.8-2.4T-A95B 71.2
- 8 Claude Opus 5 70.8
- 9 Kimi K3 70.3
- 10
Gemini 3.6 Flash 70.1
- 11 Claude Sonnet 5 69.4
- 12 ByteDance Seed: Seed 2.1 Turbo 69.2
- 13 OpenAI: GPT-5.6 Luna Pro 69.2
- 14
Tencent Hy3 68.9
- 15
DeepSeek V4 Pro 0813 68.5
- 16 Meta: Muse Glimmer 30B 68.4
- 17 GLM-5.2 66.9
- 18
MiniMax-M3 66.7
- 19 NVIDIA: Nemotron 3.5 Lightning 65.2
- 20
Inkling 64.5
- 21
DeepSeek V4 Flash 0731 62.7
- 22 Mistral: Mistral Large 3 2512 61.4
- 23
Llama 4 Maverick 60.2
- 24
Command A+ 58.1
- 25 Cohere: Command R+ (08-2024) 36.7
Source: Official FlavourBench Powered v43 leaderboard
- 73.4 Grok 4.6
- 72.4
Muse Spark 1.2
- 72.0 GPT-5.6 Sol Pro
- 71.4
Qwen 3.8 Max
- 71.2
Qwen3.8-2.4T-A95B
- 70.8 Claude Opus 5
- 70.3 Kimi K3
- 70.1
Gemini 3.6 Flash
- 69.4 Claude Sonnet 5
- 68.9
Tencent Hy3
- 68.5
DeepSeek V4 Pro 0813
- 66.9 GLM-5.2
- 66.7
MiniMax-M3
- 64.5
Inkling
- 62.7
DeepSeek V4 Flash 0731
- 60.2
Llama 4 Maverick
- 58.1
Command A+
- 1 Grok 4.6 73.4
- 2
Muse Spark 1.2 72.4
- 3 GPT-5.6 Sol Pro 72.0
- 4
Qwen 3.8 Max 71.4
- 5
Qwen3.8-2.4T-A95B 71.2
- 6 Claude Opus 5 70.8
- 7 Kimi K3 70.3
- 8
Gemini 3.6 Flash 70.1
- 9 Claude Sonnet 5 69.4
- 10
Tencent Hy3 68.9
- 11
DeepSeek V4 Pro 0813 68.5
- 12 GLM-5.2 66.9
- 13
MiniMax-M3 66.7
- 14
Inkling 64.5
- 15
DeepSeek V4 Flash 0731 62.7
- 16
Llama 4 Maverick 60.2
- 17
Command A+ 58.1
Source: Official FlavourBench Powered v43 leaderboard
| # | Entry | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Grok 4.6 | 73.4 ±2.1 | Aug 2026 | Official leaderboard | |
| 2 | Google: Gemini 3.1 Pro Preview | 72.9 ±2.1 | Official leaderboard | ||
| 3 | | 72.4 ±2.2 | Aug 2026 | Official leaderboard | |
| 4 | GPT-5.6 Sol Pro | 72.0 ±2.2 | Jul 2026 | Official leaderboard | |
| 5 | OpenAI: GPT-5.6 Terra Pro | 71.5 ±2.2 | Official leaderboard | ||
| 6 | | 71.4 ±2.2 | Aug 2026 | Official leaderboard | |
| 7 | | 71.2 ±2.2 | Aug 2026 | Official leaderboard | |
| 8 | Claude Opus 5 | 70.8 ±2.2 | Jul 2026 | Official leaderboard | |
| 9 | Kimi K3 | 70.3 ±2.2 | Jul 2026 | Official leaderboard | |
| 10 | | 70.1 ±2.2 | Jul 2026 | Official leaderboard | |
| 11 | Claude Sonnet 5 | 69.4 ±2.2 | Jun 2026 | Official leaderboard | |
| 12 | ByteDance Seed: Seed 2.1 Turbo | 69.2 ±2.2 | Official leaderboard | ||
| 13 | OpenAI: GPT-5.6 Luna Pro | 69.2 ±2.2 | Official leaderboard | ||
| 14 | | 68.9 ±2.2 | Jul 2026 | Official leaderboard | |
| 15 | | 68.5 ±2.4 | Aug 2026 | Official leaderboard | |
| 16 | Meta: Muse Glimmer 30B | 68.4 ±2.2 | Official leaderboard | ||
| 17 | GLM-5.2 | 66.9 ±2.2 | Jun 2026 | Official leaderboard | |
| 18 | | 66.7 ±2.2 | Jun 2026 | Official leaderboard | |
| 19 | NVIDIA: Nemotron 3.5 Lightning | 65.2 ±2.3 | Official leaderboard | ||
| 20 | | 64.5 ±2.4 | Jul 2026 | Official leaderboard | |
| 21 | | 62.7 ±2.3 | Jul 2026 | Official leaderboard | |
| 22 | Mistral: Mistral Large 3 2512 | 61.4 ±2.3 | Official leaderboard | ||
| 23 | | 60.2 ±2.3 | Apr 2025 | Official leaderboard | |
| 24 | | 58.1 ±2.2 | May 2026 | Official leaderboard | |
| 25 | Cohere: Command R+ (08-2024) | 36.7 ±2 | Official leaderboard |
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Grok 4.6 | 73.4 ±2.1 | Aug 2026 | Official leaderboard | |
| 2 | | 72.4 ±2.2 | Aug 2026 | Official leaderboard | |
| 3 | GPT-5.6 Sol Pro | 72.0 ±2.2 | Jul 2026 | Official leaderboard | |
| 4 | | 71.4 ±2.2 | Aug 2026 | Official leaderboard | |
| 5 | | 71.2 ±2.2 | Aug 2026 | Official leaderboard | |
| 6 | Claude Opus 5 | 70.8 ±2.2 | Jul 2026 | Official leaderboard | |
| 7 | Kimi K3 | 70.3 ±2.2 | Jul 2026 | Official leaderboard | |
| 8 | | 70.1 ±2.2 | Jul 2026 | Official leaderboard | |
| 9 | Claude Sonnet 5 | 69.4 ±2.2 | Jun 2026 | Official leaderboard | |
| 10 | | 68.9 ±2.2 | Jul 2026 | Official leaderboard | |
| 11 | | 68.5 ±2.4 | Aug 2026 | Official leaderboard | |
| 12 | GLM-5.2 | 66.9 ±2.2 | Jun 2026 | Official leaderboard | |
| 13 | | 66.7 ±2.2 | Jun 2026 | Official leaderboard | |
| 14 | | 64.5 ±2.4 | Jul 2026 | Official leaderboard | |
| 15 | | 62.7 ±2.3 | Jul 2026 | Official leaderboard | |
| 16 | | 60.2 ±2.3 | Apr 2025 | Official leaderboard | |
| 17 | | 58.1 ±2.2 | May 2026 | 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 Constraint, Cultural composition and Pairing. Grok 4.6 leads 2 of them, and Claude Opus 5 leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.
Constraint
17 models we track have a score here.
- 1 Grok 4.6 56.9
- 2
Muse Spark 1.2 55.6
- 3
Qwen 3.8 Max ↑1 53.4
- 4
Qwen3.8-2.4T-A95B ↑1 50.9
- 5 GPT-5.6 Sol Pro ↓2 50.7
- 6
Gemini 3.6 Flash ↑2 50.2
- 7 Kimi K3 48.3
- 8
DeepSeek V4 Pro 0813 ↑3 45.9
- 9
Tencent Hy3 ↑1 45.5
- 10 Claude Opus 5 ↓4 45.3
Cultural composition
17 models we track have a score here.
- 1 Grok 4.6 99.0
- 2 GPT-5.6 Sol Pro ↑1 98.8
- 3
Muse Spark 1.2 ↓1 98.8
- 4
Gemini 3.6 Flash ↑4 98.7
- 5 Claude Sonnet 5 ↑4 98.7
- 6
Qwen3.8-2.4T-A95B ↓1 98.6
- 7 Claude Opus 5 ↓1 98.6
- 8
Tencent Hy3 ↑2 98.5
- 9
Qwen 3.8 Max ↓5 98.2
- 10 Kimi K3 ↓3 98.1
Pairing
17 models we track have a score here.
- 1 Claude Opus 5 ↑5 71.1
- 2 Grok 4.6 ↓1 70.6
- 3 GPT-5.6 Sol Pro 70.5
- 4 Kimi K3 ↑3 69.9
- 5
Qwen3.8-2.4T-A95B 69.6
- 6 Claude Sonnet 5 ↑3 69.1
- 7
Muse Spark 1.2 ↓5 68.1
- 8
Qwen 3.8 Max ↓4 67.1
- 9
DeepSeek V4 Pro 0813 ↑2 66.4
- 10
Gemini 3.6 Flash ↓2 66.3
Substitution
17 models we track have a score here.
- 1 Claude Opus 5 ↑5 68.0
- 2 GPT-5.6 Sol Pro ↑1 67.9
- 3
Qwen 3.8 Max ↑1 67.0
- 4
Muse Spark 1.2 ↓2 66.9
- 5 Grok 4.6 ↓4 66.9
- 6 Claude Sonnet 5 ↑3 66.1
- 7
Tencent Hy3 ↑3 66.1
- 8
Qwen3.8-2.4T-A95B ↓3 65.6
- 9
Gemini 3.6 Flash ↓1 65.1
- 10
DeepSeek V4 Pro 0813 ↑1 65.1
3 models have held number one on the official leaderboard since its release in August 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official FlavourBench Powered v43 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 FlavourBench Powered v43 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 Writing and taste
- Design Bench (Moonshot) In use Models make graphic designs from direct prompts or few-shot reference images. An automated grader rates the results as Elo scores. Kimi K3.1 Pro 957
- WritingBench Fading Models complete 1,000 real-world writing requests across 6 domains and 100 subdomains. A judge model scores each reply on five query-specific criteria. GPT-5 83.9
- Creative Writing (MiniMax in-house) Only in system cards Models write creative pieces for real Hailuo AI user requests. Automatic assessments score the writing.
MiniMax-Text-01 81.3%
- DeepSeek Chinese functional writing evaluation Only in system cards Models answer 3,170 everyday Chinese writing prompts such as reports, emails and speeches. Pairwise judging against a baseline model gives a win rate.
DeepSeek-V4-Pro 62.6%
- Slide generation human evaluation (Z.ai) Only in system cards People compare HTML slide decks from two models on content, layout and visual quality. Metric is overall win rate. GLM-5 67.5%
- Task-dependent output homogenization evaluation Only in system cards Models answer prompts several times. A judge scores functional diversity where it is wanted and consistency where one answer is right.
Muse Spark 0.21
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (25 entries).
- Pass Models scored 17 models we track, plus 8 other entries on its leaderboard.
- Pass Newest score The newest is from August 2026, for DeepSeek V4 Pro 0813.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 5 (#11).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 17 rows for models we track and 8 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.