# Qwen3 Max Thinking

> Source: Benchize (https://benchize.com) — https://benchize.com/models/qwen3-max-thinking. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/qwen3-max-thinking.

Alibaba

Canonical URL: https://benchize.com/models/qwen3-max-thinking
Last updated: Thu, 22 Oct 2026 07:10:12 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Alibaba

## Benchmarks

| Benchmark | Score | Rank | Source |  |
| --- | --- | --- | --- | --- |
| [τ-bench airline (legacy)](https://benchize.com/benchmarks/sierra%3Alegacy%3Aairline.md) | 69% | 3 / 14 | [sierra](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 69 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_2` \| 65.33333333333333 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_3` \| 63.5 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_4` \| 62 \| percent \| consistent_success_across_k_trials \| Higher is better \| ### Configuration benchmarkVersion v0.1.3 harness tau-bench methodology {"evaluation_date":"2026-01-23","notes":"Evaluation conducted using standard tau2-bench protocol","tau2_bench_version":"v0.1.3","user_simulator":"gpt-4.1-2025-04-14","verification":{"details":"Complete evaluation with all standard tasks","modified_prompts":false,"omitted_questions":false}} modality text Reasoning effort unknown retrieval Not reported sourceSubmission qwen3-max_qwen_2026-01-23 ### Source [Original source](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) Source sierra Original model label Qwen3-Max-Thinking Evaluated 2026-01-26 Observed 2026-10-21T16:55:04.021Z Evidence SHA-256 `eb480b51b464175389262778510061ed96bf8f4e6ae2b05eb80d3d2455473b5a` Source locator `/results/airline` |  |  |  |  |
| [SVGBench Arena](https://benchize.com/benchmarks/svgbench%3Aarena.md) | 1,485 | 73 / 191 | [svgbench](https://svgbench.ai/api/leaderboard?shape=2) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1485 \| rating \| reported \| Higher is better \| ### Configuration Reasoning effort unknown sourceModelId model_qwen_qwen3_max_thinking ### Reported statistics coverageCount 27 draws 1 judged 89 losses 40 rank 98 ratingDeviation 28 wins 48 ### Source [Original source](https://svgbench.ai/api/leaderboard?shape=2) Source svgbench Original model label Qwen3 Max Thinking Observed 2026-10-22T04:30:04.538Z Evidence SHA-256 `5c0a9c869593820d57d2c05f6d416e1b44a740b4835e66c66e9ad8b8edc59e2c` Source locator `/rows/94` |  |  |  |  |
| [τ-bench retail (legacy)](https://benchize.com/benchmarks/sierra%3Alegacy%3Aretail.md) | 79.39% | 8 / 14 | [sierra](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 79.3859649122807 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_2` \| 69.44444444444446 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_3` \| 62.71929824561403 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_4` \| 57.89473684210527 \| percent \| consistent_success_across_k_trials \| Higher is better \| ### Configuration benchmarkVersion v0.1.3 harness tau-bench methodology {"evaluation_date":"2026-01-23","notes":"Evaluation conducted using standard tau2-bench protocol","tau2_bench_version":"v0.1.3","user_simulator":"gpt-4.1-2025-04-14","verification":{"details":"Complete evaluation with all standard tasks","modified_prompts":false,"omitted_questions":false}} modality text Reasoning effort unknown retrieval Not reported sourceSubmission qwen3-max_qwen_2026-01-23 ### Source [Original source](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) Source sierra Original model label Qwen3-Max-Thinking Evaluated 2026-01-26 Observed 2026-10-21T16:55:04.021Z Evidence SHA-256 `eb480b51b464175389262778510061ed96bf8f4e6ae2b05eb80d3d2455473b5a` Source locator `/results/retail` |  |  |  |  |
| [τ-bench telecom (legacy)](https://benchize.com/benchmarks/sierra%3Alegacy%3Atelecom.md) | 98.25% | 1 / 10 | [sierra](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 98.24561403508771 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_2` \| 96.63742690058479 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_3` \| 95.17543859649122 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_4` \| 93.85964912280701 \| percent \| consistent_success_across_k_trials \| Higher is better \| ### Configuration benchmarkVersion v0.1.3 harness tau-bench methodology {"evaluation_date":"2026-01-23","notes":"Evaluation conducted using standard tau2-bench protocol","tau2_bench_version":"v0.1.3","user_simulator":"gpt-4.1-2025-04-14","verification":{"details":"Complete evaluation with all standard tasks","modified_prompts":false,"omitted_questions":false}} modality text Reasoning effort unknown retrieval Not reported sourceSubmission qwen3-max_qwen_2026-01-23 ### Source [Original source](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/qwen3-max_qwen_2026-01-23/submission.json) Source sierra Original model label Qwen3-Max-Thinking Evaluated 2026-01-26 Observed 2026-10-21T16:55:04.021Z Evidence SHA-256 `eb480b51b464175389262778510061ed96bf8f4e6ae2b05eb80d3d2455473b5a` Source locator `/results/telecom` |  |  |  |  |

---
Cite as: Qwen3 Max Thinking — Benchize, https://benchize.com/models/qwen3-max-thinking (retrieved 2026-10-22)