Undominated.ai

Cheaper alternatives to Qwen3.5-122B-A10B

Qwen LMArena 1417.9 $0.715/M on a balanced workload prices as of 2026-08-28

1 model is both better and cheaper, giving up nothing.

Of the 133 models carrying both an independent score and a published price at the same delivery mode, 24 score at least as high as Qwen3.5-122B-A10B and cost no more under at least one workload. 1 matches or beats it on every capability we hold — context, maximum output, input modes, tool use and extended reasoning. 23 do not, and every row names what it drops.

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 from that comparison and nothing else. When no model is both better and cheaper than Qwen3.5-122B-A10B, it is not generated.

Where it holds

WorkloadQwen3.5-122B-A10B $/MBetter and cheaperWith a trade
Balanced$0.715114
Summarise$0.351115
Chat$0.988120
Code gen$1.35119
Agentic$0.533121

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

Intelligence1417.9
Context window262K
Max output236K
Input modestext, image, video
Tool useyes
Extended reasoningyes

A replacement has to clear every line above, not just the score. Full record for Qwen3.5-122B-A10B.

Better and cheaper, nothing given up 1

Each of these matches or beats Qwen3.5-122B-A10B on context, maximum output, input modes, tool use and reasoning, scores at least as high, and costs no more.

#ModelIntelligence$/MHolds under
1MiniMax M3 MiniMax1434.8 +16.9$0.525
  • Balanced −27%
  • Summarise −21%
  • Chat −40%
  • Code gen −39%
  • Agentic −45%

Cheaper and higher-scoring, but you give something up 23

These score at least as high and cost no more on the two plotted axes, and lose something that is not on them. Read the last column before switching.

ModelIntelligence$/MHolds underWhat you give up
MiMo-V2.5-Pro Xiaomi1465 +47.1$0.544
  • Balanced −24%
  • Summarise −5%
  • Chat −51%
  • Code gen −51%
  • Agentic −54%
  • 236K → 131K max output
  • no image, video input
Qwen3.7 Plus Qwen1456.2 +38.3$0.560
  • Balanced −22%
  • Summarise −16%
  • Chat −37%
  • Code gen −35%
  • Agentic −42%
  • 236K → 131K max output
  • no video input
Gemma 4 31B Google1441.7 +23.8$0.152
  • Balanced −79%
  • Summarise −74%
  • Chat −82%
  • Code gen −82%
  • Agentic −81%
  • 236K → 16K max output
Hy3 Tencent1441.2 +23.3$0.144
  • Balanced −80%
  • Summarise −78%
  • Chat −84%
  • Code gen −83%
  • Agentic −84%
  • 236K → 128K max output
  • no image, video input
Gemma 4 26B A4B Google1434.6 +16.7$0.138
  • Balanced −81%
  • Summarise −76%
  • Chat −82%
  • Code gen −83%
  • Agentic −79%
  • 236K → 16K max output
DeepSeek V4 Flash 0423 DeepSeek1431.6 +13.7$0.101
  • Balanced −86%
  • Summarise −81%
  • Chat −90%
  • Code gen −91%
  • Agentic −90%
  • no image, video input
MiMo-V2.5 Xiaomi1427.3 tie$0.175
  • Balanced −76%
  • Summarise −69%
  • Chat −84%
  • Code gen −84%
  • Agentic −85%
  • 236K → 131K max output
GPT-5.6 Luna OpenAI1428.5 tie$0.450
  • Balanced −37%
  • Summarise −43%
  • Chat −45%
  • Code gen −42%
  • Agentic −54%
  • 236K → 128K max output
  • no video input
DeepSeek V3.2 DeepSeek1424.6 tie$0.290
  • Balanced −59%
  • Summarise −35%
  • Chat −73%
  • Code gen −76%
  • Agentic −62%
  • 262K → 164K context
  • 236K → 147K max output
  • no image, video input
DeepSeek V3.2 Exp DeepSeek1424.4 tie$0.305
  • Balanced −57%
  • Summarise −21%
  • Chat −67%
  • Code gen −74%
  • Agentic −45%
  • 262K → 164K context
  • 236K → 66K max output
  • no image, video input
Qwen3 235B A22B Instruct 2507 Qwen1419.3 tie$0.205
  • Balanced −71%
  • Summarise −68%
  • Chat −72%
  • Code gen −73%
  • Agentic −70%
  • 236K → 16K max output
  • no image, video input
  • no extended reasoning
Qwen3 Next 80B A3B Instruct Qwen1418.6 tie$0.350
  • Balanced −51%
  • Summarise −60%
  • Chat −50%
  • Code gen −48%
  • Agentic −56%
  • no image, video input
  • no extended reasoning
DeepSeek V3.1 Terminus DeepSeek1419.6 tie$0.453
  • Balanced −37%
  • Summarise −24%
  • Chat −47%
  • Code gen −48%
  • Agentic −44%
  • 262K → 164K context
  • 236K → 33K max output
  • no image, video input
Qwen3 VL 235B A22B Instruct Qwen1420.9 tie$0.632
  • Balanced −12%
  • Summarise −25%
  • Chat −14%
  • Code gen −10%
  • Agentic −25%
  • 236K → 33K max output
  • no video input
  • no extended reasoning
Gemini 3.7 Flash Google1490.2 +72.3$0.874 chat
  • Chat −12%
  • Code gen −8%
  • Agentic −25%
  • 236K → 66K max output
Kimi K2.5 Moonshot AI1445.2 +27.3$0.494 agentic
  • Agentic −7%
  • no video input
GLM 5 Z.ai1445.2 +27.3$0.984 chat
  • Chat under 1% less
  • Agentic −4%
  • 262K → 205K context
  • 236K → 128K max output
  • no image, video input
GLM 4.6 Z.ai1439.8 +21.9$0.980 chat
  • Chat −1%
  • Agentic −9%
  • 262K → 205K context
  • 236K → 131K max output
  • no image, video input
DeepSeek V4 Pro 0423 DeepSeek1439.2 +21.3$0.979 chat
  • Chat −1%
  • Code gen −2%
  • Agentic −1%
  • no image, video input
Gemini 3.5 Flash Lite Google1436.5 +18.6$0.333 summarise
  • Summarise −5%
  • Agentic −12%
  • 236K → 66K max output
GLM 4.7 Z.ai1435.3 +17.4$0.844 chat
  • Chat −15%
  • Code gen −12%
  • Agentic −23%
  • 262K → 205K context
  • 236K → 131K max output
  • no image, video input
Qwen3.6 Plus Qwen1436.8 +18.9$0.975 chat
  • Chat −1%
  • Code gen −4%
  • 236K → 66K max output
DeepSeek V3.1 DeepSeek1419.1 tie$1.21 code gen
  • Code gen −11%
  • 262K → 164K context
  • 236K → 145K max output
  • no image, video input

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 ±10.68 points, so a gap smaller than that is marked tie rather than an improvement — see significance bands.
  • 188 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.
  • The models that nothing beats on both axes are on the value frontier, and every other model something cheaper beats is listed here.