Agents · Benchmark profile
A broad research-agent benchmark for open-ended information gathering, synthesis, and answer construction across wide search spaces.
Data verified 21 Jul 2026 · Methodology 1.6.0
Benchmark score on WideResearch
Kimi K2.6 leads at 80.8%, followed by Claude Opus 4.5 (76.4%) and Qwen3.6 Plus (74.3%).
Visual analysis
Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.
7 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| Kimi K2.6 | #1 | Moonshot AI | 80.8% |
| Claude Opus 4.5 | #2 | Anthropic | 76.4% |
| Qwen3.6 Plus | #3 | Alibaba Cloud | 74.3% |
| Qwen3.5 397B | #4 | Alibaba Cloud | 74% |
| Kimi K2.5 | #5 | Moonshot AI | 72.7% |
| GLM-5 | #6 | Z.AI | 69.8% |
| Qwen3.6-35B-A3B | #7 | Alibaba Cloud | 60.1% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | Kimi K2.6 kimi-k2.6 | Moonshot AI | open | Estimated reference | 80.8% |
| #2 | Claude Opus 4.5 claude-opus-4-5 | Anthropic | closed | Reference only | 76.4% |
| #3 | Qwen3.6 Plus qwen3-6-plus | Alibaba Cloud | closed | Reference only | 74.3% |
| #4 | Qwen3.5 397B qwen3.5-397b-a17b | Alibaba Cloud | open | Estimated reference | 74% |
| #5 | Kimi K2.5 kimi-k2.5 | Moonshot AI | — | Reference only | 72.7% |
| #6 | GLM-5 glm-5 | Z.AI | open | Reference only | 69.8% |
| #7 | Qwen3.6-35B-A3B qwen3-6-35b-a3b | Alibaba Cloud | open | Estimated reference | 60.1% |
The top of this snapshot is led by Kimi K2.6 at 80.8%; third place is 6.5 points behind. The top-7 spread is 20.7 points.
About WideResearch
A broad research-agent benchmark for open-ended information gathering, synthesis, and answer construction across wide search spaces. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A broad research-agent benchmark for open-ended information gathering, synthesis, and answer construction across wide search spaces.
Kimi K2.6 by Moonshot AI currently leads with 80.8%.
7 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
