Best Value Multimodal AI Model in 2026 — Cost-Adjusted Rankings

Multimodal workloads — processing images, charts, documents, and screenshots — often involve large inputs that drive up token costs quickly. This ranking divides each model's weighted multimodal score (MMMU-Pro, OfficeQA-Pro) by output token price. For document processing pipelines and visual AI applications running at scale, the value leaders here offer the best multimodal reasoning per dollar spent.

According to BenchLM.ai, Gemini 3.1 Flash-Lite leads this ranking with a score of 182.75, followed by GPT-4.1 nano (148.25) and Gemini 2.5 Flash (112.75). There is a significant gap between the leading models and the rest of the field.

The best open-weight option is DeepSeek Coder 2.0 (ranked #5 with a score of 53.23). While proprietary models lead, open-weight options are within striking distance for teams willing to trade a few points of performance for full model control.

This ranking is based on weighted averages across the scoring benchmarks in multimodalGrounded tracked by BenchLM.ai. For detailed model profiles, click any model name below. To compare two specific models head-to-head, use the "vs #" links.

Gemini 3.1 Flash-Lite
GoogleProprietary1M

182.75

Score/$

Score: 73.1 · $0.4/1M

GPT-4.1 nano
OpenAIProprietary1M

148.25

Score/$

Score: 59.3 · $0.4/1M

Gemini 2.5 Flash
GoogleProprietary1M

112.75

Score/$

Score: 67.7 · $0.6/1M

4
GPT-4o mini
OpenAIProprietary128K

100.25

Score/$

Score: 60.2 · $0.6/1M

5
DeepSeek Coder 2.0
DeepSeekOpen Weight128K

53.23

Score/$

Score: 58.6 · $1.1/1M

6
GPT-5.4 nano
OpenAIProprietary400K

52.88

Score/$

Score: 66.1 · $1.25/1M

7
GPT-4.1 mini
OpenAIProprietary1M

46.25

Score/$

Score: 74 · $1.6/1M

8
Gemini 3 Flash
GoogleProprietary1M

26.52

Score/$

Score: 79.6 · $3/1M

9
Kimi K2.5
Moonshot AIOpen Weight128K

26.51

Score/$

Score: 74.2 · $2.8/1M

10
DeepSeek-R1
DeepSeekOpen Weight128K

21.69

Score/$

Score: 47.5 · $2.19/1M

11
Claude Haiku 4.5
AnthropicProprietary200K

19.6

Score/$

Score: 78.4 · $4/1M

12
Gemini 3.1 Pro
GoogleProprietary1M

19

Score/$

Score: 95 · $5/1M

13
Gemini 2.5 Pro
GoogleProprietary1M

17.02

Score/$

Score: 85.1 · $5/1M

14
o3-mini
OpenAIProprietary200K

16.9

Score/$

Score: 74.4 · $4.4/1M

15
GPT-5.1
OpenAIProprietary200K

15.29

Score/$

Score: 91.8 · $6/1M

16
Mistral Large 3
MistralProprietary128K

12.58

Score/$

Score: 75.5 · $6/1M

17
GPT-5.1-Codex-Max
OpenAIProprietary400K

11.02

Score/$

Score: 88.2 · $8/1M

18
GPT-5.2-Codex
OpenAIProprietary400K

10.95

Score/$

Score: 87.6 · $8/1M

19
GPT-5.2
OpenAIProprietary400K

10.81

Score/$

Score: 86.5 · $8/1M

20
GPT-4.1
OpenAIProprietary1M

9.2

Score/$

Score: 73.6 · $8/1M

21
GPT-5.3 Codex
OpenAIProprietary400K

9.13

Score/$

Score: 91.3 · $10/1M

22
GPT-4o
OpenAIProprietary128K

7

Score/$

Score: 70 · $10/1M

23
Claude Sonnet 4.5
AnthropicProprietary200K

6.33

Score/$

Score: 95 · $15/1M

24
Grok 4.1
xAIProprietary1M

6.21

Score/$

Score: 93.2 · $15/1M

25
Claude Sonnet 4.6
AnthropicProprietary200K

6.12

Score/$

Score: 91.9 · $15/1M

26
GPT-5.4
OpenAIProprietary1.05M

5.86

Score/$

Score: 87.9 · $15/1M

27
o3
OpenAIProprietary200K

1.81

Score/$

Score: 72.3 · $40/1M

28
o1
OpenAIProprietary200K

1.18

Score/$

Score: 70.7 · $60/1M

29
Claude Opus 4.6
AnthropicProprietary1M

1.13

Score/$

Score: 84.8 · $75/1M

30
GPT-5.4 Pro
OpenAIProprietary1.05M

0.53

Score/$

Score: 94.9 · $180/1M

31
o1-pro
OpenAIProprietary200K

0.08

Score/$

Score: 48.5 · $600/1M

Key Takeaways

  • The best value model is Gemini 3.1 Flash-Lite by Google with a Score/$ ratio of 182.75 (score: 73.1, output: $0.4/1M tokens).
  • The best open-weight model in this ranking is DeepSeek Coder 2.0 at position #5.
  • 31 models are included in this ranking.
Last updated: April 3, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.