AI Battle > Rankings > MMMU-Pro: Total Cost to Run

MMMU-Pro: Total Cost to Run Ranking — eval mmmu (2026)

Gemini 3.7 Flash (low) by Google leads the MMMU-Pro: Total Cost to Run ranking with a score of 1.68. The ranking covers 19 evaluated eval mmmu from a public dataset that AI Battle re-checks daily.

Last updated: . Source: public benchmark datasets and provider pricing, aggregated daily; 19 entries shown.

Gemini 3.7 Flash (low)1.68Muse Glimmer (high)0.36Gemini 3.5 Flash-Lite0.67GPT-5.6 Luna (max)0.59Gemini 3.7 Flash (high)1.68MiniMax-M30.41DeepSeek V4.1 Flash (Reasoning, Max Effort)0.25Qwen3.8 27B (xhigh)0.49GPT-6 Astra (medium)10.0Gemini 3.8 Flash (high)1.68GPT-6 Astra (high)10.0Inkling (xhigh)1.76Mistral Medium 3.50.50GPT-6 Astra (xhigh)10.0GPT-5.6 Sol (max)4.00GPT-5.6 Terra (max)3.18GPT-6 Astra (max)10.0Claude Opus 5 (Adaptive Reasoning, Max Effor6.04Kimi K3 (max)3.63
Every entry in the dataset, scaled from the smallest to the largest published value.

All 19 entries in this dataset

#EntryCreatorValueΔ vs #1
1Gemini 3.7 Flash (low)Google1.680.0%
2Muse Glimmer (high)Meta0.36-78.7%
3Gemini 3.5 Flash-LiteGoogle0.67-60.0%
4GPT-5.6 Luna (max)OpenAI0.59-64.9%
5Gemini 3.7 Flash (high)Google1.680.0%
6MiniMax-M3MiniMax0.41-75.8%
7DeepSeek V4.1 Flash (Reasoning, Max Effort)DeepSeek0.25-85.1%
8Qwen3.8 27B (xhigh)Alibaba0.49-71.2%
9GPT-6 Astra (medium)OpenAI10.0493.8%
10Gemini 3.8 Flash (high)Google1.680.0%
11GPT-6 Astra (high)OpenAI10.0493.8%
12Inkling (xhigh)Thinking Machines1.764.7%
13Mistral Medium 3.5Mistral0.50-70.6%
14GPT-6 Astra (xhigh)OpenAI10.0493.8%
15GPT-5.6 Sol (max)OpenAI4.00137.2%
16GPT-5.6 Terra (max)OpenAI3.1888.9%
17GPT-6 Astra (max)OpenAI10.0493.8%
18Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic6.04258.3%
19Kimi K3 (max)Kimi3.63115.6%

How the values are distributed

Across the 19 numeric entries in this window the median is 1.68 and the mean is 3.52, so the typical entry sits 0.00 below the leader. The window spans 0.25 to 10.0.

74211004
8 equal buckets between the smallest (0.25) and largest (10.0) value; numbers under bars are entry counts.

12 of 19 entries are at or above the median, and the top three hold 45% of the combined value — read it against the metric below, since a higher number is not "better" for every dataset.

Creators in this ranking

10 creators appear in this window; the three most-represented hold 63% of the entries.

CreatorEntriesHighest valueTop entry
OpenAI710.0GPT-6 Astra (medium)
Google41.68Gemini 3.7 Flash (low)
Anthropic16.04Claude Opus 5 (Adaptive Reasoning, Max Effort)
Kimi13.63Kimi K3 (max)
Thinking Machines11.76Inkling (xhigh)
Mistral10.50Mistral Medium 3.5
Alibaba10.49Qwen3.8 27B (xhigh)
MiniMax10.41MiniMax-M3
Meta10.36Muse Glimmer (high)
DeepSeek10.25DeepSeek V4.1 Flash (Reasoning, Max Effort)

Where these models appear in other rankings

The same models are measured across multiple independent datasets. Aggregating the 19 entries above against every other AI Battle dataset gives this cross-section, computed by AI Battle and published nowhere upstream:

ModelDatasetsBest placements elsewhere
Gemini 3.7 Flash (low)3#5/19 in MMMU-Pro: Token Usage, #7/19 in MMMU-Pro: Score
Muse Glimmer (high)43#1/20 in Output Tokens per Intelligence Index Task, #1/20 in Cost per Intelligence Index Task, #1/20 in Cost per Intelligence Index Task
Gemini 3.5 Flash-Lite44#1/20 in End-to-End Response Time, #1/20 in Output Speed, #1/20 in Time per Intelligence Index Task
GPT-5.6 Luna (max)58#1/11 in Cost per Task, #1/11 in Cost per Task, #3/20 in Pricing: Cache Hit, Input, and Output
Gemini 3.7 Flash (high)3#5/19 in MMMU-Pro: Score, #7/19 in MMMU-Pro: Token Usage
MiniMax-M357#1/20 in AA-Omniscience Hallucination Rate, #1/20 in 𝜏³-Banking: Output Tokens per Task, #4/20 in Pricing: Cache Hit, Input, and Output
DeepSeek V4.1 Flash (Reasoning, Max Effort)56#3/20 in End-to-End Response Time, #3/20 in Latency: Time To First Answer Token, #2/11 in Cost per Task
Qwen3.8 27B (xhigh)43#6/20 in AA-Omniscience Hallucination Rate, #6/20 in 𝜏³-Banking: Score, #8/20 in AA-LCR v1.1: Cost per Task
GPT-6 Astra (medium)12#1/20 in AA-Omniscience Index: Token Usage, #1/19 in MMMU-Pro: Token Usage, #2/20 in Terminal-Bench v4.0: Output Tokens per Task
Gemini 3.8 Flash (high)60#1/11 in Speed, #1/11 in Speed, #2/20 in Output Speed
GPT-6 Astra (high)12#1/20 in AA-Omniscience Index: Score, #2/19 in MMMU-Pro: Token Usage, #2/19 in MMMU-Pro: Score
Inkling (xhigh)52#3/20 in Context Window, #3/20 in 𝜏³-Banking: Output Tokens per Task, #5/20 in AA-Briefcase Output Tokens per Task
Mistral Medium 3.541#2/20 in GDPval-AA v2: Output Tokens per Task, #3/20 in Output Tokens per Intelligence Index Task, #3/20 in Output Tokens per Intelligence Index Task
GPT-6 Astra (xhigh)15#1/20 in Terminal-Bench v4.0: Score, #2/20 in Humanity's Last Exam: Output Tokens per Task, #3/20 in AA-Omniscience Index: Score
GPT-5.6 Sol (max)56#3/20 in SciCode: Output Tokens per Task, #4/20 in 𝜏³-Banking: Output Tokens per Task, #5/20 in AA-LCR v1.1: Output Tokens per Task
GPT-5.6 Terra (max)52#7/20 in AA-Briefcase Output Tokens per Task, #8/20 in Humanity's Last Exam: Output Tokens per Task, #9/20 in Output Speed
GPT-6 Astra (max)51#1/19 in MMMU-Pro: Score, #2/20 in AA-Omniscience Index, #2/20 in Artificial Analysis Intelligence Index by Open Weights / Proprietary
Claude Opus 5 (Adaptive Reasoning, Max Effort)44#2/20 in AA-Briefcase Elo, #2/20 in GDPval-AA v2 Leaderboard, #2/20 in AA-Briefcase Elo
Kimi K3 (max)60#1/20 in Context Window, #1/20 in AA-LCR v1.1: Score, #1/14 in Model Size: Total and Active Parameters

Metric reference

This ranking carries the metrics answerCost, inputCost, reasoningCost from the MMMU-Pro: Total Cost to Run dataset in the tab::eval-mmmu family. Units and the direction of "better" are not declared by the source dataset, so AI Battle does not invent them: values are shown exactly as published, and the ordering is the source dataset's own.

How this ranking is built

Rebuilt on from the public dataset MMMU-Pro: Total Cost to Run (family tab::eval-mmmu). The source lists 19 entries and all 19 are shown here. AI Battle requests the dataset's rows, keeps the ordering the source publishes, and computes only the statistics on this page: median, mean, spread, the #1–#2 gap, the creator breakdown and the raw-data file. Nothing on this page is generated text about the entries — the numbers come from the source.

Refresh cadence: every deploy re-pulls the dataset, so this page tracks the source's own update cycle. Corrections: see the methodology page.

Limitations

Raw data

Machine-readable copy of this table, with the statistics below and the source provenance: /data/rankings/eval-mmmu-mmmu-pro-total-cost-to-run.json. Free to reuse with attribution to AI Battle and the source dataset.

FAQ — MMMU-Pro: Total Cost to Run

What is the best model for MMMU-Pro: Total Cost to Run in 2026?

Gemini 3.7 Flash (low) by Google currently leads the MMMU-Pro: Total Cost to Run ranking with a score of 1.68, the highest value in this dataset across eval mmmu. AI Battle aggregates the dataset from its public source and re-checks it daily.

How often is the MMMU-Pro: Total Cost to Run ranking updated?

The MMMU-Pro: Total Cost to Run ranking is rebuilt daily from the upstream public dataset. Scores are the source dataset's own published values, so positions track whatever that source last released — each ranking page names the dataset it comes from.

Which models rank in the top 3 for MMMU-Pro: Total Cost to Run?

The top 3 are: 1. Gemini 3.7 Flash (low) (1.68), 2. Muse Glimmer (high) (0.36), 3. Gemini 3.5 Flash-Lite (0.67).

All AI Battle Rankings

Directories