LOADING COMPARISON
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
PICK MODELS TO COMPARE
Now pick a second model to put against DeepSeek V3.1 Terminus. Fair fights only, so these are all chat and coding models.
SELECTED: DEEPSEEK V3.1 TERMINUS
Grok 4.6MULTIMODAL#1 · 47,328 VOICES
7.7VIBE
Qwen3.8 27BMULTIMODAL#2 · 14,896 VOICES
7.6VIBE- DiscoveryTEXT
#3 · 11,946 VOICES
7.6VIBE - Fugu MaxMULTIMODAL
#4 · 5,959 VOICES
7.6VIBE - Schematron V2 TurboTEXT
#5 · 5,871 VOICES
7.6VIBE 
Muse Spark 1.3TEXT#6 · 9,188 VOICES
7.5VIBE- Qwen3.8 FlashMULTIMODAL
#7 · 6,385 VOICES
7.5VIBE - Ternary Bonsai 2 27BMULTIMODAL
#9 · 4,353 VOICES
7.5VIBE - Mercury 2.5TEXT
#11 · 6,877 VOICES
7.4VIBE - Step 1TEXT
#13 · 4,033 VOICES
7.4VIBE - Dots3 NoteMULTIMODAL
#16 · 2,641 VOICES
7.4VIBE - Apertus v1.5 70BTEXT
#17 · 2,527 VOICES
7.4VIBE