Cheaper alternatives to GPT-4o-mini (2024-07-18)

OpenAI LMArena 1286.6 $0.263/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, 25 score at least as high as GPT-4o-mini (2024-07-18) 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. 24 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-4o-mini (2024-07-18), it is not generated.

Where it holds

Workload	GPT-4o-mini (2024-07-18) $/M	Better and cheaper	With a trade
Balanced	$0.263	1	20
Summarise	$0.151	1	21
Chat	$0.307	1	21
Code gen	$0.414	1	22
Agentic	$0.173	1	18

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

Intelligence	1286.6
Context window	128K
Max output	16K
Input modes	text, image, file
Tool use	yes
Extended reasoning	no

A replacement has to clear every line above, not just the score. Full record for GPT-4o-mini (2024-07-18).

Better and cheaper, nothing given up 1

Each of these matches or beats GPT-4o-mini (2024-07-18) on context, maximum output, input modes, tool use and reasoning, scores at least as high, and costs no more.

#	Model	Intelligence	$/M	Holds under
1	GPT-5 Nano OpenAI	1320.3 +33.7	$0.138	Balanced −48% Summarise −64% Chat −43% Code gen −38% Agentic −56%

Cheaper and higher-scoring, but you give something up 24

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.

Model	Intelligence	$/M	Holds under	What you give up
Gemma 4 31B Google	1441.7 +155.1	$0.152	Balanced −42% Summarise −40% Chat −42% Code gen −43% Agentic −40%	no file input
Hy3 Tencent	1441.2 +154.6	$0.144	Balanced −45% Summarise −49% Chat −47% Code gen −45% Agentic −52%	no image, file input
Gemma 4 26B A4B Google	1434.6 +148	$0.138	Balanced −48% Summarise −45% Chat −42% Code gen −44% Agentic −36%	no file input
DeepSeek V4 Flash 0423 DeepSeek	1431.6 +145	$0.101	Balanced −61% Summarise −56% Chat −69% Code gen −70% Agentic −68%	no image, file input
MiMo-V2.5 Xiaomi	1427.3 +140.7	$0.175	Balanced −33% Summarise −29% Chat −50% Code gen −49% Agentic −54%	no file input
Qwen3 235B A22B Instruct 2507 Qwen	1419.3 +132.7	$0.205	Balanced −22% Summarise −25% Chat −11% Code gen −12% Agentic −8%	no image, file input
Step 3.5 Flash StepFun	1403.8 +117.2	$0.150	Balanced −43% Summarise −27% Chat −41% Code gen −47% Agentic −25%	no image, file input
Qwen3.5-Flash Qwen	1397.6 +111	$0.114	Balanced −57% Summarise −51% Chat −53% Code gen −56% Agentic −45%	no file input
Qwen3 30B A3B Instruct 2507 Qwen	1384.3 +97.7	$0.084	Balanced −68% Summarise −63% Chat −65% Code gen −67% Agentic −60%	no image, file input
Solar Pro 4 Upstage	1376.2 +89.6	$0.052	Balanced −80% Summarise −82% Chat −81% Code gen −80% Agentic −83%	no image, file input
gpt-oss-120b OpenAI	1365.6 +79	$0.070	Balanced −73% Summarise −71% Chat −71% Code gen −72% Agentic −67%	no image, file input
Gemma 3 27B Google	1358.3 +71.7	$0.172	Balanced −34% Summarise −42% Chat −30% Code gen −28% Agentic −35%	no file input
GLM 4.7 Flash Z.ai	1352.9 +66.3	$0.145	Balanced −45% Summarise −58% Chat −41% Code gen −37% Agentic −53%	no image, file input
Qwen3 32B Qwen	1340.1 +53.5	$0.130	Balanced −50% Summarise −40% Chat −48% Code gen −52% Agentic −36%	no image, file input
Gemma 3 12B Google	1334.2 +47.6	$0.075	Balanced −71% Summarise −64% Chat −71% Code gen −73% Agentic −62%	no file input
Qwen3 30B A3B Qwen	1316.9 +30.3	$0.215	Balanced −18% Summarise −8% Chat −12% Code gen −16%	no image, file input
Olmo 3 32B Think Allen AI	1298.5 +11.9	$0.237	Balanced −10% Chat −6% Code gen −13%	128K → 66K context no image, file input no tool use
Granite 4.1 8B IBM	1291.6 tie	$0.063	Balanced −76% Summarise −65% Chat −77% Code gen −81% Agentic −67%	no image, file input
Gemma 3 4B Google	1290.8 tie	$0.063	Balanced −76% Summarise −65% Chat −77% Code gen −81% Agentic −67%	no file input no tool use
gpt-oss-20b OpenAI	1287.8 tie	$0.055	Balanced −79% Summarise −77% Chat −77% Code gen −78% Agentic −74%	no image, file input
DeepSeek V3.2 DeepSeek	1424.6 +138	$0.269 chat	Chat −13% Code gen −22%	no image, file input
DeepSeek V3.2 Exp DeepSeek	1424.4 +137.8	$0.354 code gen	Code gen −14%	no image, file input
Qwen3 Next 80B A3B Instruct Qwen	1418.6 +132	$0.141 summarise	Summarise −6%	no image, file input
GLM 4.5 Air Z.ai	1382.8 +96.2	$0.136 summarise	Summarise −10%	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.