Instructions following: Did not follow instructions
Instructions following
Did not follow instructions
See which AI models are most likely to hit Did not follow instructions on Instructions following, so you can spot weak points faster.
Failure Reasons
48/48
Filter models
No models match the current search and filters.
| Rank | Model | Company | Did not follow instructions Count | Category Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #26 | Muse Spark 1.3 high | Meta | 1 | 6.5 | $1.653 | 1/2 | 14.7s |
| #38 | Muse Spark 1.1 medium | Meta | 1 | 6.5 | $1.420 | 1/2 | 6.31s |
| #51 | Muse Spark 1.1 low | Meta | 1 | 7.3 | $0.673 | 1/2 | 5.42s |
| #52 | Claude Fable 5.1 medium | Anthropic | 1 | 7.3 | $2.909 | 1/2 | 5.77s |
| #58 | Muse Glimmer 30B xhigh | Meta | 1 | 8.4 | $0.510 | 1/2 | 8.90s |
| #60 | Muse Spark 1.1 high | Meta | 1 | 6.4 | $2.253 | 1/2 | 7.81s |
| #83 | DeepSeek V4 Pro high | DeepSeek | 1 | 7.8 | $0.489 | 1/2 | 8.73s |
| #87 | Solar Pro 4 xhigh | Upstage | 1 | 8.4 | $0.050 | 1/2 | 51.3s |
| #118 | Muse Glimmer 30B high | Meta | 1 | 8.4 | $0.430 | 1/2 | 8.62s |
| #120 | Qwen3.7 Flash low | Qwen | 1 | 8.5 | $0.043 | 1/2 | 9.88s |
| #123 | Muse Glimmer 30B low | Meta | 1 | 8.4 | $0.152 | 1/2 | 3.60s |
| #125 | Solar Pro 4 high | Upstage | 1 | 7.9 | $0.046 | 1/2 | 41.5s |
| #127 | GPT-5.6 Sol none | OpenAI | 1 | 8.5 | $0.201 | 1/2 | 1.33s |
| #137 | Qwen3.7 Flash medium | Qwen | 1 | 5.8 | $0.065 | 0/2 | 9.27s |
| #143 | Solar Pro 4 low | Upstage | 1 | 8.0 | $0.044 | 1/2 | 56.4s |