Undominated.ai

Cheaper alternatives to GPT-5 Nano

OpenAI LMArena 1320.3 $0.138/M on a balanced workload prices as of 2026-08-28

Everything cheaper and higher-scoring gives something up.

Of the 133 models carrying both an independent score and a published price at the same delivery mode, 12 score at least as high as GPT-5 Nano and cost no more under at least one workload. 12 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 GPT-5 Nano, it is not generated.

Where it holds

WorkloadGPT-5 Nano $/MBetter and cheaperWith a trade
Balanced$0.13808
Summarise$0.05502
Chat$0.17609
Code gen$0.256012
Agentic$0.07605

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

Intelligence1320.3
Context window400K
Max output128K
Input modestext, image, file
Tool useyes
Extended reasoningyes

A replacement has to clear every line above, not just the score. Full record for GPT-5 Nano.

Cheaper and higher-scoring, but you give something up 12

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
Gemma 4 26B A4B Google1434.6 +114.3$0.138
  • Balanced same price
  • Code gen −10%
  • 400K → 262K context
  • 128K → 16K max output
  • no file input
DeepSeek V4 Flash 0423 DeepSeek1431.6 +111.3$0.101
  • Balanced −26%
  • Chat −47%
  • Code gen −51%
  • Agentic −28%
  • no image, file input
Qwen3.5-Flash Qwen1397.6 +77.3$0.114
  • Balanced −17%
  • Chat −19%
  • Code gen −29%
  • 128K → 66K max output
  • no file input
Qwen3 30B A3B Instruct 2507 Qwen1384.3 +64$0.084
  • Balanced −39%
  • Chat −40%
  • Code gen −47%
  • Agentic −8%
  • 400K → 262K context
  • 128K → 32K max output
  • no image, file input
  • no extended reasoning
Solar Pro 4 Upstage1376.2 +55.9$0.052
  • Balanced −62%
  • Summarise −49%
  • Chat −67%
  • Code gen −68%
  • Agentic −61%
  • no image, file input
gpt-oss-120b OpenAI1365.6 +45.3$0.070
  • Balanced −49%
  • Summarise −20%
  • Chat −49%
  • Code gen −54%
  • Agentic −25%
  • 400K → 131K context
  • 128K → 118K max output
  • no image, file input
Qwen3 32B Qwen1340.1 +19.8$0.130
  • Balanced −5%
  • Chat −9%
  • Code gen −22%
  • 400K → 131K context
  • 128K → 16K max output
  • no image, file input
Gemma 3 12B Google1334.2 +13.9$0.075
  • Balanced −45%
  • Chat −49%
  • Code gen −57%
  • Agentic −14%
  • 400K → 131K context
  • 128K → 16K max output
  • no file input
  • no extended reasoning
Gemma 4 31B Google1441.7 +121.4$0.237 code gen
  • Code gen −8%
  • 400K → 262K context
  • 128K → 16K max output
  • no file input
Hy3 Tencent1441.2 +120.9$0.163 chat
  • Chat −8%
  • Code gen −12%
  • 400K → 262K context
  • no image, file input
MiMo-V2.5 Xiaomi1427.3 +107$0.155 chat
  • Chat −12%
  • Code gen −17%
  • no file input
Step 3.5 Flash StepFun1403.8 +83.5$0.220 code gen
  • Code gen −14%
  • 400K → 262K context
  • 128K → 66K max output
  • no image, file 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.