AI Battle > Rankings > πΒ³-Banking: Output Tokens per Task
MiniMax-M3 by MiniMax leads the πΒ³-Banking: Output Tokens per Task ranking with a score of 2468. The ranking covers 20 evaluated eval tau3 from a public dataset that AI Battle re-checks daily.
Last updated: . Source: public benchmark datasets and provider pricing, aggregated daily; 20 entries shown.
| # | Entry | Creator | Value | Ξ vs #1 |
|---|---|---|---|---|
| 1 | MiniMax-M3 | MiniMax | 2468 | 0.0% |
| 2 | Muse Glimmer (high) | Meta | 3419 | 38.5% |
| 3 | Inkling (xhigh) | Thinking Machines | 2278 | -7.7% |
| 4 | GPT-5.6 Sol (max) | OpenAI | 2405 | -2.5% |
| 5 | GLM-5.3 (max) | Z AI | 2546 | 3.1% |
| 6 | GPT-5.6 Luna (max) | OpenAI | 2075 | -15.9% |
| 7 | Qwen3.8 27B (medium) | Alibaba | 2883 | 16.8% |
| 8 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) | Anthropic | 4587 | 85.9% |
| 9 | Nemotron 3 Ultra 550B A55B (Reasoning) | NVIDIA | 3294 | 33.5% |
| 10 | Qwen3.8 Max | Alibaba | 2888 | 17.0% |
| 11 | Qwen3.8 27B (xhigh) | Alibaba | 2966 | 20.2% |
| 12 | Kimi K3 (max) | Kimi | 3497 | 41.7% |
| 13 | Qwen3.8 2.4T A95B | Alibaba | 2794 | 13.2% |
| 14 | GPT-5.6 Terra (max) | OpenAI | 2188 | -11.4% |
| 15 | Gemini 3.5 Flash-Lite | 1925 | -22.0% | |
| 16 | Grok 4.6 (high) | SpaceXAI | 2794 | 13.2% |
| 17 | Mistral Medium 3.5 | Mistral | 2300 | -6.8% |
| 18 | Muse Spark 1.3 (xhigh) | Meta | 4261 | 72.6% |
| 19 | GLM-5.3-Flash | Z AI | 2589 | 4.9% |
| 20 | GPT-6 Astra (max) | OpenAI | 2189 | -11.3% |
Across the 20 numeric entries in this window the median is 2691 and the mean is 2817, so the typical entry sits 223 above the leader. The window spans 1925 to 4587.
10 of 20 entries are at or above the median, and the top three hold 22% of the combined value β read it against the metric below, since a higher number is not "better" for every dataset.
12 creators appear in this window; the three most-represented hold 50% of the entries.
| Creator | Entries | Highest value | Top entry |
|---|---|---|---|
| Alibaba | 4 | 2966 | Qwen3.8 27B (xhigh) |
| OpenAI | 4 | 2405 | GPT-5.6 Sol (max) |
| Meta | 2 | 4261 | Muse Spark 1.3 (xhigh) |
| Z AI | 2 | 2589 | GLM-5.3-Flash |
| Anthropic | 1 | 4587 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) |
| Kimi | 1 | 3497 | Kimi K3 (max) |
| NVIDIA | 1 | 3294 | Nemotron 3 Ultra 550B A55B (Reasoning) |
| SpaceXAI | 1 | 2794 | Grok 4.6 (high) |
| MiniMax | 1 | 2468 | MiniMax-M3 |
| Mistral | 1 | 2300 | Mistral Medium 3.5 |
| Thinking Machines | 1 | 2278 | Inkling (xhigh) |
| 1 | 1925 | Gemini 3.5 Flash-Lite |
The same models are measured across multiple independent datasets. Aggregating the 20 entries above against every other AI Battle dataset gives this cross-section, computed by AI Battle and published nowhere upstream:
| Model | Datasets | Best placements elsewhere |
|---|---|---|
| MiniMax-M3 | 57 | #1/20 in AA-Omniscience Hallucination Rate, #4/20 in Pricing: Cache Hit, Input, and Output, #4/20 in Pricing: Cache Hit, Input, and Output |
| 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 |
| Inkling (xhigh) | 52 | #3/20 in Context Window, #5/20 in AA-Briefcase Output Tokens per Task, #5/20 in GDPval-AA v2: Output Tokens per Task |
| GPT-5.6 Sol (max) | 56 | #3/20 in SciCode: Output Tokens per Task, #5/20 in AA-LCR v1.1: Output Tokens per Task, #6/20 in Artificial Analysis Intelligence Index by Open Weights / Proprietary |
| GLM-5.3 (max) | 57 | #3/20 in AA-Briefcase Tool Calls Breakdown, Avg per Task, #4/20 in GDPval-AA v2 Leaderboard, #4/20 in AA-Omniscience Hallucination Rate |
| 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 |
| Qwen3.8 27B (medium) | 3 | #8/20 in πΒ³-Banking: Score, #9/20 in πΒ³-Banking: Cost per Task |
| Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) | 20 | #2/20 in Humanity's Last Exam: Score, #2/20 in AA-Omniscience Accuracy, #2/20 in AA-Briefcase Rubric Pass Rate by File Type (Normalized) |
| Nemotron 3 Ultra 550B A55B (Reasoning) | 46 | #1/12 in Artificial Analysis Openness Index: Score, #1/12 in Artificial Analysis Openness Index: Components, #2/20 in πΒ³-Banking: Time per Task |
| Qwen3.8 Max | 3 | #1/20 in πΒ³-Banking: Score, #20/20 in πΒ³-Banking: Cost per Task |
| Qwen3.8 27B (xhigh) | 43 | #6/20 in AA-Omniscience Hallucination Rate, #6/20 in πΒ³-Banking: Score, #8/20 in AA-LCR v1.1: Cost per Task |
| 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 |
| Qwen3.8 2.4T A95B | 43 | #2/14 in Model Size: Total and Active Parameters, #5/20 in AA-Briefcase Tool Calls Breakdown, Avg per Task, #5/20 in πΒ³-Banking: Score |
| 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 |
| Gemini 3.5 Flash-Lite | 44 | #1/20 in End-to-End Response Time, #1/20 in Output Speed, #1/20 in Time per Intelligence Index Task |
| Grok 4.6 (high) | 61 | #2/20 in πΒ³-Banking: Score, #5/20 in AA-Briefcase Elo, #5/20 in AA-Omniscience Index |
| Mistral Medium 3.5 | 41 | #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 |
| Muse Spark 1.3 (xhigh) | 12 | #3/20 in πΒ³-Banking: Time per Task, #5/20 in SciCode: Score, #7/20 in GDPval-AA v2 Leaderboard |
| GLM-5.3-Flash | 54 | #1/20 in Pricing: Cache Hit, Input, and Output, #1/20 in Pricing: Cache Hit, Input, and Output, #1/20 in AA-Omniscience Index: Cost Breakdown |
| 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 |
This ranking carries the metrics
answer, reasoning from the
πΒ³-Banking: Output Tokens per Task dataset in the tab::eval-tau3 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.
Rebuilt on from the
public dataset πΒ³-Banking: Output Tokens per Task (family tab::eval-tau3).
The source lists 20 entries and all 20 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 #1β#2 gap, the creator breakdown and the raw-data file.
Nothing on this page is generated text about the entries β 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.
Machine-readable copy of this table, with the statistics below and the source provenance: /data/rankings/eval-tau3-3-banking-output-tokens-per-task.json. Free to reuse with attribution to AI Battle and the source dataset.
MiniMax-M3 by MiniMax currently leads the πΒ³-Banking: Output Tokens per Task ranking with a score of 2468, the highest value in this dataset across eval tau3. AI Battle aggregates the dataset from its public source and re-checks it daily.
The πΒ³-Banking: Output Tokens per Task 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 β each ranking page names the dataset it comes from.
The top 3 are: 1. MiniMax-M3 (2468), 2. Muse Glimmer (high) (3419), 3. Inkling (xhigh) (2278).