xbench-DeepSearch
Step 3.5 Flash 83.7% leads since Feb 2026 Fading Agents search the web to answer 100 Chinese questions that need multi-step retrieval. Scored by accuracy against reference answers.
A refreshed 100-question set, DeepSearch-2510, followed in October 2025.
- Level
- Expert Specialists with years of training, often over hours of work
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Agent task Many steps in a terminal, browser, desktop, app, game or robot.
Step 3.5 Flash leads with 83.7% among models released by Feb 2026
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Feb 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Step 3.5 Flash 83.7%
- 2 Grok 4 Fast 74%
- 3
MiniMax-M2 72%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Step 3.5 Flash | 83.7% | Feb 2026 | Model card | |
| 2 | Grok 4 Fast | 74% | Sep 2025 | Model card | |
| 3 | MiniMax-M2 | 72% | Oct 2025 | Model card |
2 models have held number one since its release in May 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs headline it less often now, last in Feb 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 2 Feb 2026 | Step 3.5 Flash | 83.7 | xbench-DeepSearch (2025.05) |
| 27 Oct 2025 | MiniMax-M2 | 72 | xbench-DeepSearch |
| 19 Sep 2025 | Grok 4 Fast | 74 | X Bench Deepsearch (zh) pass@1 |
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 11 Feb 2026 | Step 3.5 Flash technical report | 83.7 | xbench-DeepSearch-2505, avg@3 (Table 5) | Capabilities |
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
More in Tool use
- AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state.
Gemini 4 Carbon 53.9% - MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks.
Muse Spark 1.1 88.1% - BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes.
Qwen3.7-Max 75% - τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
- Claw-Eval In use 300 human-verified agent tasks (service orchestration, multimodal, multi-turn dialogue) audited via traces and snapshots. Rubric scores with Pass^3 consistency. Claude Sonnet 4.6 81.4%
- DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%