How does Qwen3.6-35B-A3B compare with other LLMs?

AskClash compares Qwen3.6-35B-A3B against nearby AI models using public benchmark scores, pricing, context window, and access details.

What benchmarks are tracked for Qwen3.6-35B-A3B?

The page shows cached public benchmark cells such as HLE, GPQA, SWE-bench, SWE-Pro, Terminal-Bench, MCP Atlas, MMMU-Pro, ARC-AGI-2, Tau2, and related model scores when available.

LLM Leaderboard · Open Weight

Qwen3.6-35B-A3B benchmarks, pricing, and LLM comparison.

Compare Qwen3.6-35B-A3B vs GPT, Claude, Gemini, DeepSeek, open-weight, and frontier AI models using public benchmark scores, token pricing, context window, and access details.

Open live LLM leaderboard Open in app

Rank #31AskClash overall score: 38.6

$0 / $0Input and output token price, when published. Context: 262K.

API/OAuthBilling and access path cached for this model row.

Qwen3.6-35B-A3B benchmark snapshot

AskClash combines public LLM benchmark cells into a weighted percentile score and penalizes missing coverage so narrow rows do not dominate better-measured models.

Overall38.6

Benchmark cells9

Context262K

CreatorAlibaba

Qwen3.6-35B-A3B public benchmark scores

Cached benchmark values can include HLE, GPQA, SWE-bench, SWE-Pro, SWE-Atlas, Terminal-Bench, MCP Atlas, MMMU-Pro, ARC-AGI-2, Tau2, and model-specific coding or agent scores.

HLE

21.4 score

GPQA

86.0 score

SWE-bench

73.4 score

Terminal-Bench

51.5 score

LiveCodeBench

80.4 score

MCP Atlas

62.8 score

CharXiv

78.0 score

MMMU-Pro

75.3 score

Tau2

95.3 score

Qwen3.6-35B-A3B vs other AI models

Use these comparison links to evaluate Qwen3.6-35B-A3B against nearby LLMs by benchmark score, price, context window, and provider.

Claude Mythos/Fable 5 vs Qwen3.6-35B-A3B Claude Opus 4.8 (Adaptive) vs Qwen3.6-35B-A3B GPT-5.5 xHigh vs Qwen3.6-35B-A3B GLM-5.2 vs Qwen3.6-35B-A3B Claude Opus 4.7 (Adaptive) vs Qwen3.6-35B-A3B Qwen3.7 Max vs Qwen3.6-35B-A3B GPT-5.4 xHigh vs Qwen3.6-35B-A3B Kimi K2.7 Code vs Qwen3.6-35B-A3B

Related AI and tech coverage

Cached AskClash article matches that can provide release, provider, benchmark, pricing, or market context around this model.

Qwen3.6-35B-A3B on my laptop drew me a better pelican than Claude Opus 4.7

Simon Willison

HuggingFace TRL / RLHF v1.3.0 Release Notes

TRL v1.3 ships training support for the new **Qwen 3.6** family (`Qwen/Qwen3.6-27B`, `Qwen/Qwen3.6-35B-A3B`). Qwen 3.6 reuses the `Qwen3_5Moe*` architecture but ships a slightly different chat template (adds a `preserve_thinking` flag, tweaks tool-arg stringification), so exact-string template matching needed updates across the stack. A new experimental `TPOTrainer` implements [Triple Preference Optimization](https://huggingface.co/papers/2405.16681), which augments DPO with a `reference` (gold)

Sources: Alibaba replaces DingTalk CEO Chen Hang with Chen Yusen after an internal debate over the enterprise communication app's role in Alibaba's AI strategy (Bloomberg)

Techmeme

Sources: a new AI lab founded by Junyang Lin, former lead of Alibaba's Qwen models, has raised several hundred million dollars at a ~$2B post-money valuation (The Information)

Techmeme

Last cached leaderboard date: June 18, 2026. This model page is generated from the AskClash LLM Leaderboard cache and linked from the live leaderboard.