Our new X account is live! Follow @wizwand_team for updates

Mathematical Multimodal Reasoning on MathVista (Accuracy)

85.6Accuracy

Seed-1.5-thinking

Updated 4d ago

Evaluation Results

Method	Links
Seed-1.5-thinking 2025.12		85.6
Gemini-2.5-Pro-Thinking 2025.12		83.8
InternVL-3.5-4B 2026.02		77.1
DenseMLLM-4B 2026.02		76.5
Vision-G1 2025.12		76.1
PeBR-R1 2025.12		76
VAPO-Thinker 2025.12		75.6
ThinkLite-VL 2025.12		75.1
VL-Rethinker 2025.12		74.9
Qwen2.5-VL-72B 2025.12		74.8
Qwen3-VL-4B 2026.02		73.7
VisonR1 7B 2026.01		73.5
Ours 2025.12		73.4
Revisual-R1 2025.12		73.1
MM-Eureka 2025.12		73
MathBook 2025.12		73
OpenVLThinker 2025.12		72.3
Look-Back 2025.12		72.3
GPT-4.1 2025.12		72
InternVL3-8B 2025.12		71.6
MMR1-Math-v0 2025.12		71
MixedR1 7B 2026.01		70.6
Claude-4-Sonnet 2025.12		70.4
CodeDance-7B 2025.12		70.3
ARM-Thinker-7B 2025.12		70.2
Ours [with DPS and annealing] 2026.01		70.1
DeepEyes-7B 2025.12		70.1
Ours [with DPS] 2026.01		70
Ours 2026.01		68.7
Qwen2.5-VL 2025.12		68.2
Qwen2.5-VL-7B 2025.12		68.1
VLAA-Thinker 7B 2026.01		68
VLAA-Thinking 2025.12		68
Qwen2.5-VL-7B 2025.12		67.8
Qwen2.5VL 7B 2026.01		65.3
InternVL2.5 8B 2026.01		64.4
InternVL2.5-8B 2025.12		64.4
R1-Onevision 7B 2026.01		64.1
R1-Onevision-7B 2025.12		64.1
R1-VL-7B 2025.12		63.5
GPT-4o 2025.12		63.4
LLaVA-OneVision 7B 2026.01		63.2
Llava-OV-7B 2025.12		63.2
Mulberry 2026.01		63.1
Gemma-3-27B 2025.12		59.8
GPT-4V 2026.01		49.9