AI Battle > Rankings > Mean Turns per Task
Qwen3.8-Flash-Next by Alibaba leads the Mean Turns per Task ranking with a score of 242. The ranking covers 20 evaluated eval briefcase 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 | Δ vs #1 |
|---|---|---|---|---|
| 1 | Qwen3.8-Flash-Next | Alibaba | 242 | 0.0% |
| 2 | GLM-5.3-Flash | Z AI | 147 | -39.5% |
| 3 | gpt-oss-120b (high) | OpenAI | 113 | -53.4% |
| 4 | GLM-5.3 (max) | Z AI | 110 | -54.6% |
| 5 | DeepSeek V4.1 Flash (Reasoning, Max Effort) | DeepSeek | 109 | -55.0% |
| 6 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | Anthropic | 105 | -56.6% |
| 7 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | Anthropic | 103 | -57.5% |
| 8 | Qwen3.8 2.4T A95B | Alibaba | 101 | -58.4% |
| 9 | MiniMax-M3 | MiniMax | 101 | -58.4% |
| 10 | Gemini 3.8 Flash (high) | 99.6 | -58.8% | |
| 11 | Muse Spark 1.3 (max) | Meta | 99.1 | -59.0% |
| 12 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | Anthropic | 90.9 | -62.5% |
| 13 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) | Anthropic | 89.8 | -62.9% |
| 14 | K2 Horizon 375B A23B | MBZUAI Institute of Foundation Models | 84.2 | -65.2% |
| 15 | Kimi K3 (max) | Kimi | 82.6 | -65.9% |
| 16 | Qwen3.8 27B (xhigh) | Alibaba | 82.3 | -66.0% |
| 17 | Inkling (xhigh) | Thinking Machines | 81.5 | -66.3% |
| 18 | Claude Opus 5 (Adaptive Reasoning, High Effort) | Anthropic | 76.3 | -68.5% |
| 19 | GPT-5.6 Luna (max) | OpenAI | 72.7 | -69.9% |
| 20 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | Anthropic | 67.1 | -72.3% |
Across the 20 numeric entries in this window the median is 99.3 and the mean is 103, so the typical entry sits 143 below the leader. The window spans 67.1 to 242.
10 of 20 entries are at or above the median, and the top three hold 24% of the combined value — read it against the metric below, since a higher number is not "better" for every dataset.
11 creators appear in this window; the three most-represented hold 55% of the entries.
| Creator | Entries | Highest value | Top entry |
|---|---|---|---|
| Anthropic | 6 | 105 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) |
| Alibaba | 3 | 242 | Qwen3.8-Flash-Next |
| Z AI | 2 | 147 | GLM-5.3-Flash |
| OpenAI | 2 | 113 | gpt-oss-120b (high) |
| DeepSeek | 1 | 109 | DeepSeek V4.1 Flash (Reasoning, Max Effort) |
| MiniMax | 1 | 101 | MiniMax-M3 |
| 1 | 99.6 | Gemini 3.8 Flash (high) | |
| Meta | 1 | 99.1 | Muse Spark 1.3 (max) |
| MBZUAI Institute of Foundation Models | 1 | 84.2 | K2 Horizon 375B A23B |
| Kimi | 1 | 82.6 | Kimi K3 (max) |
| Thinking Machines | 1 | 81.5 | Inkling (xhigh) |
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 |
|---|---|---|
| Qwen3.8-Flash-Next | 5 | #1/20 in AA-Briefcase Tool Calls Breakdown, Avg per Task, #6/20 in AA-Briefcase Elo, #8/20 in AA-Briefcase Rubric Pass Rate by File Type (Normalized) |
| 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-oss-120b (high) | 39 | #1/20 in Cost to Run Artificial Analysis Intelligence Index, #1/20 in Cost to Run Artificial Analysis Intelligence Index, #1/20 in Humanity's Last Exam: Cost per Task |
| 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 |
| 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 |
| Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | 36 | #1/20 in AA-Briefcase Elo, #1/20 in AA-Omniscience Index, #1/20 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 |
| 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 𝜏³-Banking: Score |
| MiniMax-M3 | 57 | #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 |
| Gemini 3.8 Flash (high) | 60 | #1/11 in Speed, #1/11 in Speed, #2/20 in Output Speed |
| Muse Spark 1.3 (max) | 61 | #3/20 in AA-Briefcase Elo, #3/20 in Output Speed, #3/20 in GDPval-AA v2 Leaderboard |
| Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | 11 | #4/20 in AA-Briefcase Rubric Pass Rate by File Type (Normalized), #4/20 in AA-Briefcase Elo, #4/20 in GDPval-AA v2 Leaderboard |
| 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) |
| K2 Horizon 375B A23B | 24 | #2/20 in AA-Omniscience Hallucination Rate, #8/20 in AA-Briefcase Tool Calls Breakdown, Avg per Task, #12/20 in AA-Omniscience Index: Token Usage |
| 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 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 |
| 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 |
| Claude Opus 5 (Adaptive Reasoning, High Effort) | 6 | #9/20 in AA-Briefcase Rubric Pass Rate by File Type (Normalized), #9/20 in AA-Briefcase Elo, #15/20 in Time per 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 |
| Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | 40 | #2/20 in SciCode: Score, #3/20 in AA-Omniscience Index, #3/20 in AA-Omniscience Index |
This ranking carries the metric
aaBriefcaseMeanTurnsPerTask from the
Mean Turns per Task dataset in the tab::eval-briefcase 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 Mean Turns per Task (family tab::eval-briefcase).
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 #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.
Machine-readable copy of this table, with the statistics below and the source provenance: /data/rankings/eval-briefcase-mean-turns-per-task.json. Free to reuse with attribution to AI Battle and the source dataset.
Qwen3.8-Flash-Next by Alibaba currently leads the Mean Turns per Task ranking with a score of 242, the highest value in this dataset across eval briefcase. AI Battle aggregates the dataset from its public source and re-checks it daily.
The Mean Turns 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 — each ranking page names the dataset it comes from.
The top 3 are: 1. Qwen3.8-Flash-Next (242), 2. GLM-5.3-Flash (147), 3. gpt-oss-120b (high) (113).