Undominated.ai

Cheaper alternatives to GPT-4.1 Mini

OpenAI LMArena 1340.5 $0.700/M on a balanced workload prices as of 2026-08-28

5 models are 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, 42 score at least as high as GPT-4.1 Mini and cost no more under at least one workload. 5 match or beat it on every capability we hold — context, maximum output, input modes, tool use and extended reasoning. 37 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-4.1 Mini, it is not generated.

Where it holds

WorkloadGPT-4.1 Mini $/MBetter and cheaperWith a trade
Balanced$0.700235
Summarise$0.374536
Chat$0.790233
Code gen$1.10233
Agentic$0.402335

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

Intelligence1340.5
Context window1.0M
Max output33K
Input modesimage, text, file
Tool useyes
Extended reasoningno

A replacement has to clear every line above, not just the score. Full record for GPT-4.1 Mini.

Better and cheaper, nothing given up 5

Each of these matches or beats GPT-4.1 Mini on context, maximum output, input modes, tool use and reasoning, scores at least as high, and costs no more.

#ModelIntelligence$/MHolds under
1GPT-5.6 Luna OpenAI1428.5 +88$0.450
  • Balanced −36%
  • Summarise −47%
  • Chat −31%
  • Code gen −28%
  • Agentic −40%
2Gemini 3.1 Flash Lite Preview Google1414.8 +74.3$0.563
  • Balanced −20%
  • Summarise −34%
  • Chat −14%
  • Code gen −10%
  • Agentic −24%
3Gemini 3.7 Flash Google1490.2 +149.7$0.354 summarise
  • Summarise −6%
  • Agentic −1%
4Gemini 3.5 Flash Lite Google1436.5 +96$0.333 summarise
  • Summarise −11%
5Gemini 2.5 Flash Google1417.3 +76.8$0.333 summarise
  • Summarise −11%

Cheaper and higher-scoring, but you give something up 37

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 +124.5$0.544
  • Balanced −22%
  • Summarise −11%
  • Chat −39%
  • Code gen −40%
  • Agentic −39%
  • no image, file input
Qwen3.7 Plus Qwen1456.2 +115.7$0.560
  • Balanced −20%
  • Summarise −21%
  • Chat −21%
  • Code gen −20%
  • Agentic −22%
  • 1.0M → 1.0M context
  • no file input
Gemma 4 31B Google1441.7 +101.2$0.152
  • Balanced −78%
  • Summarise −76%
  • Chat −77%
  • Code gen −78%
  • Agentic −74%
  • 1.0M → 262K context
  • 33K → 16K max output
  • no file input
Hy3 Tencent1441.2 +100.7$0.144
  • Balanced −79%
  • Summarise −79%
  • Chat −79%
  • Code gen −79%
  • Agentic −79%
  • 1.0M → 262K context
  • no image, file input
Gemma 4 26B A4B Google1434.6 +94.1$0.138
  • Balanced −80%
  • Summarise −78%
  • Chat −77%
  • Code gen −79%
  • Agentic −72%
  • 1.0M → 262K context
  • 33K → 16K max output
  • no file input
DeepSeek V4 Flash 0423 DeepSeek1431.6 +91.1$0.101
  • Balanced −86%
  • Summarise −82%
  • Chat −88%
  • Code gen −89%
  • Agentic −86%
  • no image, file input
MiniMax M3 MiniMax1434.8 +94.3$0.525
  • Balanced −25%
  • Summarise −26%
  • Chat −26%
  • Code gen −25%
  • Agentic −27%
  • no file input
MiMo-V2.5 Xiaomi1427.3 +86.8$0.175
  • Balanced −75%
  • Summarise −71%
  • Chat −80%
  • Code gen −81%
  • Agentic −80%
  • no file input
DeepSeek V3.2 DeepSeek1424.6 +84.1$0.290
  • Balanced −59%
  • Summarise −39%
  • Chat −66%
  • Code gen −71%
  • Agentic −50%
  • 1.0M → 164K context
  • no image, file input
DeepSeek V3.2 Exp DeepSeek1424.4 +83.9$0.305
  • Balanced −56%
  • Summarise −26%
  • Chat −59%
  • Code gen −68%
  • Agentic −28%
  • 1.0M → 164K context
  • no image, file input
Qwen3 235B A22B Instruct 2507 Qwen1419.3 +78.8$0.205
  • Balanced −71%
  • Summarise −70%
  • Chat −65%
  • Code gen −67%
  • Agentic −60%
  • 1.0M → 262K context
  • 33K → 16K max output
  • no image, file input
Qwen3 Next 80B A3B Instruct Qwen1418.6 +78.1$0.350
  • Balanced −50%
  • Summarise −62%
  • Chat −38%
  • Code gen −36%
  • Agentic −42%
  • 1.0M → 262K context
  • no image, file input
DeepSeek V3.1 Terminus DeepSeek1419.6 +79.1$0.453
  • Balanced −35%
  • Summarise −28%
  • Chat −34%
  • Code gen −36%
  • Agentic −25%
  • 1.0M → 164K context
  • no image, file input
Qwen3 VL 235B A22B Instruct Qwen1420.9 +80.4$0.632
  • Balanced −10%
  • Summarise −30%
  • Agentic −1%
  • 1.0M → 262K context
  • no file input
Inkling Small Thinking Machines1411.7 +71.2$0.637
  • Balanced −9%
  • Chat −18%
  • Code gen −20%
  • Agentic −12%
  • no file input
Step 3.5 Flash StepFun1403.8 +63.3$0.150
  • Balanced −79%
  • Summarise −71%
  • Chat −77%
  • Code gen −80%
  • Agentic −68%
  • 1.0M → 262K context
  • no image, file input
Qwen3.5-27B Qwen1407.9 +67.4$0.536
  • Balanced −23%
  • Summarise −30%
  • Chat −6%
  • Code gen −7%
  • Agentic under 1% less
  • 1.0M → 262K context
  • no file input
MiniMax M2.7 MiniMax1405.3 +64.8$0.525
  • Balanced −25%
  • Summarise −26%
  • Chat −26%
  • Code gen −25%
  • Agentic −27%
  • 1.0M → 205K context
  • no image, file input
Qwen3.5-Flash Qwen1397.6 +57.1$0.114
  • Balanced −84%
  • Summarise −80%
  • Chat −82%
  • Code gen −83%
  • Agentic −77%
  • 1.0M → 1.0M context
  • no file input
Qwen3.5-35B-A3B Qwen1395.6 +55.1$0.500
  • Balanced −29%
  • Summarise −20%
  • Chat −18%
  • Code gen −22%
  • Agentic under 1% less
  • 1.0M → 262K context
  • no file input
Qwen3 30B A3B Instruct 2507 Qwen1384.3 +43.8$0.084
  • Balanced −88%
  • Summarise −85%
  • Chat −87%
  • Code gen −88%
  • Agentic −83%
  • 1.0M → 262K context
  • 33K → 32K max output
  • no image, file input
GLM 4.5 Air Z.ai1382.8 +42.3$0.310
  • Balanced −56%
  • Summarise −64%
  • Chat −51%
  • Code gen −49%
  • Agentic −56%
  • 1.0M → 131K context
  • no image, file input
Solar Pro 4 Upstage1376.2 +35.7$0.052
  • Balanced −93%
  • Summarise −93%
  • Chat −93%
  • Code gen −93%
  • Agentic −93%
  • 1.0M → 524K context
  • no image, file input
DeepSeek V3 0324 DeepSeek1375 +34.5$0.438
  • Balanced −37%
  • Summarise −23%
  • Chat −30%
  • Code gen −36%
  • Agentic −10%
  • 1.0M → 164K context
  • no image, file input
GLM 4.6V Z.ai1374.7 +34.2$0.450
  • Balanced −36%
  • Summarise −31%
  • Chat −41%
  • Code gen −42%
  • Agentic −39%
  • 1.0M → 131K context
  • no file input
GPT-5.4 Nano OpenAI1372.8 +32.3$0.463
  • Balanced −34%
  • Summarise −46%
  • Chat −28%
  • Code gen −26%
  • Agentic −38%
  • 1.0M → 400K context
gpt-oss-120b OpenAI1365.6 +25.1$0.070
  • Balanced −90%
  • Summarise −88%
  • Chat −89%
  • Code gen −89%
  • Agentic −86%
  • 1.0M → 131K context
  • no image, file input
GPT-5 Mini OpenAI1373.4 +32.9$0.688
  • Balanced −2%
  • Summarise −27%
  • Agentic −6%
  • 1.0M → 400K context
Qwen3 Next 80B A3B Thinking Qwen1367.5 +27$0.412
  • Balanced −41%
  • Summarise −46%
  • Chat −28%
  • Code gen −29%
  • Agentic −23%
  • 1.0M → 262K context
  • no image, file input
Gemma 3 27B Google1358.3 +17.8$0.172
  • Balanced −75%
  • Summarise −77%
  • Chat −73%
  • Code gen −73%
  • Agentic −72%
  • 1.0M → 262K context
  • no file input
Mercury 2 Inception1357.8 +17.3$0.375
  • Balanced −46%
  • Summarise −44%
  • Chat −52%
  • Code gen −51%
  • Agentic −52%
  • 1.0M → 128K context
  • no image, file input
MiniMax M2.5 MiniMax1359 +18.5$0.472
  • Balanced −33%
  • Summarise −36%
  • Chat −34%
  • Code gen −33%
  • Agentic −39%
  • 1.0M → 205K context
  • no image, file input
GLM 4.7 Flash Z.ai1352.9 +12.4$0.145
  • Balanced −79%
  • Summarise −83%
  • Chat −77%
  • Code gen −76%
  • Agentic −80%
  • 1.0M → 203K context
  • 33K → 16K max output
  • no image, file input
Trinity Large Thinking Arcee1341.9 tie$0.378
  • Balanced −46%
  • Summarise −45%
  • Chat −46%
  • Code gen −47%
  • Agentic −45%
  • 1.0M → 262K context
  • no image, file input
MiniMax M2 MiniMax1342.1 tie$0.446
  • Balanced −36%
  • Summarise −22%
  • Chat −29%
  • Code gen −35%
  • Agentic −8%
  • 1.0M → 205K context
  • no image, file input
Qwen3.5-122B-A10B Qwen1417.9 +77.4$0.351 summarise
  • Summarise −6%
  • 1.0M → 262K context
  • no file input
Qwen3 235B A22B Thinking 2507 Qwen1413.8 +73.3$0.334 summarise
  • Summarise −11%
  • 1.0M → 131K context
  • 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.