Undominated.ai

Alternatives to Qwen3.6 Max Preview

Qwen LMArena 1446.8 $2.31/M on a balanced workload prices as of 2026-10-06

17 models are higher-scoring or cheaper, with neither dimension worse.

Of the 144 models carrying both an independent score and a published price at the same delivery mode, 18 score at least as high as Qwen3.6 Max Preview and cost no more under at least one listed workload, with at least one of those dimensions improved. 17 match or beat its recorded context, maximum output, input modes, tool use and extended reasoning. 1 has a recorded capability loss, named on every row.

Dominance is not a property of a model. It is a property of a model, a capability lens and a workload, and all three are named on every row. Scores are LMArena Elo, used under CC BY 4.0 from the official dataset; prices are blended per million tokens. A higher score is not a drop-in replacement.

This page is built only while a candidate improves score or price without worsening the other under a listed workload. When that comparison no longer holds, it is not generated.

Where it holds

WorkloadQwen3.6 Max Preview $/MRecorded capabilities preservedRecorded capability losses
Balanced$2.31131
Summarise$1.28141
Chat$3.08131
Code gen$4.11131
Agentic$1.80161

Workload mixes are defined on the methodology page. Cached input is priced at the cached rate, and reasoning tokens at the worse of the reasoning and output rates.

What you would be replacing

Intelligence1446.8
Context window262K
Max output65K
Input modestext
Tool useyes
Extended reasoningyes

Preserving recorded capabilities requires matching or beating these fields. Full record for Qwen3.6 Max Preview, or check Qwen3.6 Max Preview against the whole catalogue.

Recorded capabilities preserved 17

Each option has a higher score or a lower price, with neither dimension worse under the listed workloads. It matches or beats Qwen3.6 Max Preview on recorded context, maximum output, input modes, tool use and extended reasoning.

#ModelIntelligence$/MHolds under
1MiMo-V2.6-Pro Xiaomi1491 +44.2$0.540
  • Balanced −77%
  • Summarise −74%
  • Chat −84%
  • Code gen −84%
  • Agentic −87%
2Muse Spark 1.3 Meta1489.9 +43.1$2.00
  • Balanced −13%
  • Summarise −15%
  • Chat −31%
  • Code gen −28%
  • Agentic −42%
3Muse Spark 1.2 Meta1483.2 +36.4$2.00
  • Balanced −13%
  • Summarise −15%
  • Chat −31%
  • Code gen −28%
  • Agentic −42%
4Gemini 3.6 Flash Google1479.5 +32.7$1.50
  • Balanced −35%
  • Summarise −45%
  • Chat −43%
  • Code gen −39%
  • Agentic −56%
5Muse Spark 1.1 Meta1479.2 +32.4$2.00
  • Balanced −13%
  • Summarise −15%
  • Chat −31%
  • Code gen −28%
  • Agentic −42%
6GLM 5.3 Flash Z.ai1469.6 +22.8$0.238
  • Balanced −90%
  • Summarise −90%
  • Chat −92%
  • Code gen −91%
  • Agentic −93%
7GLM 5.3 Z.ai1471.4 +24.6$1.00
  • Balanced −57%
  • Summarise −73%
  • Chat −52%
  • Code gen −48%
  • Agentic −64%
8GLM 5.2 Z.ai1470.4 +23.6$1.12
  • Balanced −52%
  • Summarise −67%
  • Chat −50%
  • Code gen −45%
  • Agentic −61%
9MiMo-V2.5-Pro Xiaomi1465 +18.2$0.544
  • Balanced −76%
  • Summarise −74%
  • Chat −84%
  • Code gen −84%
  • Agentic −86%
10DeepSeek V4.1 Flash DeepSeek1462.5 +15.7$0.182
  • Balanced −92%
  • Summarise −95%
  • Chat −92%
  • Code gen −91%
  • Agentic −94%
11MiMo-V2.6-Flash Xiaomi1456.4 tie$0.175
  • Balanced −92%
  • Summarise −92%
  • Chat −95%
  • Code gen −95%
  • Agentic −96%
12Qwen3.7 Plus Qwen1454.6 tie$0.560
  • Balanced −76%
  • Summarise −77%
  • Chat −80%
  • Code gen −79%
  • Agentic −83%
13Kimi K2.6 Moonshot AI1455.5 tie$0.961
  • Balanced −58%
  • Summarise −64%
  • Chat −63%
  • Code gen −60%
  • Agentic −70%
14Gemini 3.8 Flash Google1497 +50.2$1.60 agentic
  • Agentic −11%
15Gemini 3.7 Flash Google1487 +40.2$1.60 agentic
  • Agentic −11%
16Kimi K3 Moonshot AI1475.9 +29.1$1.23 summarise
  • Summarise −4%
17Grok 4.5 xAI1448.1 tie$1.59 agentic
  • Agentic −12%

Alternatives with recorded capability losses 1

These improve score or price without worsening the other under the listed workloads, but reduce at least one recorded capability. Read the last column before switching.

ModelIntelligence$/MHolds underRecorded capability loss
GLM 5.1 Z.ai1461.5 +14.7$1.83
  • Balanced −21%
  • Summarise −10%
  • Chat −34%
  • Code gen −35%
  • Agentic −32%
  • 262K → 205K context

What this compares, and what it leaves out

  • Quality is LMArena Elo, used under CC BY 4.0 from the official dataset. The 95% confidence interval on a difference between two scores is about ±10.47 points, so a gap smaller than that is marked tie rather than an improvement — see significance bands.
  • 204 further models at this delivery mode carry a price but no independent score. They are absent from the comparison in both directions — unrated is not a zero, and an unmeasured model is neither an alternative nor a worse buy.
  • Retired models are never offered as an alternative, and a model only competes against its own delivery mode: batch trades latency for price, so it is not a like-for-like swap.
  • Nothing here measures latency, throughput, rate limits or how a model behaves on your prompts. Two models with the same index score are not interchangeable.
  • Models with no higher-scoring or cheaper alternative that is no worse on the other axis are on the value frontier. Models with such an alternative are listed here.
Evidence & Ask
Markdown for LLMs