Undominated.ai

Best models first. Then what they cost.

131 of 143 rated, priced standard models have a higher-scoring or cheaper alternative, with neither dimension worse. They cost a median 6.5× the cheapest such alternative. 12 are Undominated.

Check a model Audit a bill Subscription vs API

361 models · 57 providers · 46 with tiered pricing · verified 2026-10-21

143 ranked · 212 unrated · 6 batch or free only
Advanced filters
Refine

Latest price move, Oct 21, 2026: Z.ai GLM 4.7 is 26.3% cheaper on the balanced blend.

Skip past model table

A model’s rank is 1 + the number of models that clearly beat it. How they work

When a cache-read rate is unlisted, blended estimates use the selected tier’s full input rate.

AI models ranked by General capability, then by price for a Balanced workload.
# 95% interval Axis 1477–1545, following the rows in view CapabilitiesVerdictSide by side
Gemini 4 ArgonGooglenothing better for less⇄

3 models share rank 2 — the benchmark can't separate them$8.00–$20.00 /M · cheapest Claude Opus 5.5

Claude Opus 5.5AnthropicPick Gemini 4 Argonsame price · +17.7⇄
Claude Fable 5.1AnthropicPick Gemini 4 Argon60% cheaper · +23.3⇄
Muse WonderMetaPick Gemini 4 Argonsame price · +29.3⇄

2 models share rank 3 — the benchmark can't separate them$10.00 /M each

Claude Opus 5AnthropicPick Gemini 4 Argon20% cheaper · +30.7⇄
Claude Opus 4.6AnthropicPick Gemini 4 Argon20% cheaper · +31.4⇄

2 models share rank 4 — the benchmark can't separate them$0.540–$3.00 /M · cheapest MiMo-V2.6-Pro

Gemini 3.8 FlashGooglenothing better for less⇄
MiMo-V2.6-ProXiaominothing better for less⇄

2 models share rank 6 — the benchmark can't separate them$2.00–$20.00 /M · cheapest Muse Spark 1.3

Claude Fable 5AnthropicPick Gemini 4 Argon60% cheaper · +42.2⇄
Muse Spark 1.3MetaPick MiMo-V2.6-Pro +0.5, within error shorter max output +173% cheaper · +0.5, within error⇄

3 models share rank 7 — the benchmark can't separate them$2.00–$10.00 /M · cheapest Muse Spark 1.2

Claude Opus 4.7AnthropicPick Gemini 4 Argon20% cheaper · +43.5⇄
Gemini 3.7 FlashGooglePick Gemini 3.8 Flashsame price · +11.6⇄
Muse Spark 1.2MetaPick Muse Spark 1.3 +4.9, within errorsame price · +4.9, within error⇄

3 models share rank 11 — the benchmark can't separate them$1.50–$4.50 /M · cheapest Gemini 3.6 Flash

Gemini 3.1 Pro PreviewGooglePick Gemini 3.8 Flash33% cheaper · +18.4⇄
Gemini 3.6 FlashGooglePick MiMo-V2.6-Pro +10.8, within error no file input64% cheaper · +10.8, within error⇄
Muse Spark 1.1MetaPick Muse Spark 1.3same price · +11.7⇄
Gemini 3.5 FlashGooglePick Gemini 3.8 Flash11% cheaper · +21.1⇄

2 models share rank 13 — the benchmark can't separate them$3.05–$4.00 /M · cheapest Kimi K3

Kimi K3Moonshot AIPick Muse Spark 1.334% cheaper · +14.2⇄
Claude Sonnet 5.5AnthropicPick Muse Spark 1.350% cheaper · +19.2⇄

2 models share rank 15 — the benchmark can't separate them$0.238–$1.00 /M · cheapest GLM 5.3 Flash

GLM 5.3Z.aiPick MiMo-V2.6-Pro shorter max output46% cheaper · +18.9⇄
GLM 5.3 FlashZ.ainothing better for less⇄
GLM 5.2Z.aiPick GLM 5.3 +1.4, within error11% cheaper · +1.4, within error⇄

4 models share rank 19 — the benchmark can't separate them$0.182–$11.25 /M · cheapest DeepSeek V4.1 Flash

GPT-5.5OpenAIPick Gemini 4 Argon smaller context29% cheaper · +66.3⇄
MiMo-V2.5-ProXiaomiPick MiMo-V2.6-Pro shorter max output1% cheaper · +25.2⇄
GLM 5.1Z.aiPick MiMo-V2.6-Pro70% cheaper · +27.7⇄
DeepSeek V4.1 FlashDeepSeeknothing better for less⇄
Claude Opus 4.8AnthropicPick Gemini 4 Argon20% cheaper · +72.1⇄
MiMo-V2.6-FlashXiaominothing better for less⇄

4 models share rank 25 — the benchmark can't separate them$0.825–$8.00 /M · cheapest Kimi K2.6

Claude Sonnet 4.6AnthropicPick Muse Spark 1.367% cheaper · +32.0⇄
Gemini 2.5 ProGooglePick Gemini 3.8 Flash13% cheaper · +41.0⇄
GPT-5.6 SolOpenAIPick Gemini 4 Argon smaller contextsame price · +76.0⇄
Kimi K2.6Moonshot AIPick GLM 5.3 Flash71% cheaper · +15.6⇄
Qwen3.7 PlusQwenPick MiMo-V2.6-Pro4% cheaper · +35.7⇄

2 models share rank 28 — the benchmark can't separate them$4.00–$5.63 /M · cheapest GPT-6.1 Sol

GPT-5.4OpenAIPick Gemini 3.8 Flash smaller context +147% cheaper · +46.5⇄
GPT-6.1 SolOpenAIPick Gemini 3.8 Flash smaller context +125% cheaper · +51.2⇄
Claude Opus 4.5AnthropicPick Gemini 4 Argon20% cheaper · +82.3⇄

3 models share rank 31 — the benchmark can't separate them$1.21–$4.50 /M · cheapest DeepSeek V4 Pro 0423

Grok 4.5xAIPick Muse Spark 1.333% cheaper · +41.8⇄
GPT-5.6 TerraOpenAIPick GPT-6.1 Sol +0.7, within error11% cheaper · +0.7, within error⇄
DeepSeek V4 Pro 0423DeepSeekPick GLM 5.317% cheaper · +26.0⇄

2 models share rank 32 — the benchmark can't separate them$0.205–$1.09 /M · cheapest Gemma 4 31B

GLM 5Z.aiPick MiMo-V2.6-Pro50% cheaper · +44.2⇄
Gemma 4 31BGooglePick MiMo-V2.6-Flash +12.3, within error18% cheaper · +12.3, within error⇄

2 models share rank 35 — the benchmark can't separate them$1.20–$20.00 /M · cheapest Kimi K2.5

Kimi K2.5Moonshot AIPick GLM 5.3 Flash80% cheaper · +26.0⇄
GPT-6 AstraOpenAIPick GPT-5.544% cheaper · +24.8⇄

6 models share rank 36 — the benchmark can't separate them$0.760–$4.00 /M · cheapest GLM 4.6

Claude Sonnet 5AnthropicPick Muse Spark 1.350% cheaper · +47.4⇄
InklingThinking MachinesPick MiMo-V2.6-Pro shorter max output69% cheaper · +49.1⇄
Qwen3.8 27BQwenPick MiMo-V2.6-Pro44% cheaper · +49.7⇄
Step 5 PreviewStepFunPick MiMo-V2.6-Pro62% cheaper · +50.3⇄
GLM 4.6Z.aiPick MiMo-V2.6-Pro29% cheaper · +50.4⇄
GLM 5V TurboZ.aiPick MiMo-V2.6-Pro72% cheaper · +52.8⇄
Hy3TencentPick DeepSeek V4.1 Flash21% cheaper · +25.1⇄

5 models share rank 40 — the benchmark can't separate them$0.128–$6.00 /M · cheapest Gemma 4 26B A4B

Claude Sonnet 4.5AnthropicPick Gemini 3.8 Flash50% cheaper · +60.2⇄
Qwen3.5 397B A17BQwenPick MiMo-V2.6-Pro38% cheaper · +52.4⇄
Qwen3.6 PlusQwenPick MiMo-V2.6-Pro26% cheaper · +53.9⇄
GLM 4.7Z.aiPick MiMo-V2.6-Pro27% cheaper · +54.8⇄
Gemma 4 26B A4B Googlenothing better for less⇄
Gemini 3.5 Flash LiteGooglePick MiMo-V2.6-Pro no file input36% cheaper · +55.6⇄

2 models share rank 43 — the benchmark can't separate them$0.485–$2.07 /M · cheapest MiniMax M3

MiniMax M3MiniMaxPick GLM 5.3 Flash51% cheaper · +37.6⇄
Mistral Large 4MistralPick Muse Spark 1.33% cheaper · +60.1⇄
DeepSeek V4 Flash 0423DeepSeeknothing better for less⇄

2 models share rank 49 — the benchmark can't separate them$0.450–$1.00 /M · cheapest GPT-5.6 Luna

GPT-5.6 LunaOpenAIPick GLM 5.3 Flash smaller context +147% cheaper · +40.4⇄
GLM 4.5Z.airetires Dec 31, 2026Pick MiMo-V2.6-Pro46% cheaper · +60.3⇄

2 models share rank 50 — the benchmark can't separate them$0.920–$3.00 /M · cheapest R1 0528

Grok 4.6xAIPick Muse Spark 1.333% cheaper · +61.6⇄
R1 0528DeepSeekPick MiMo-V2.6-Pro41% cheaper · +62.6⇄
MiMo-V2.5XiaomiPick MiMo-V2.6-Flash shorter max output4% cheaper · +27.4⇄
DeepSeek V3.2 ExpDeepSeekPick GLM 5.3 Flash22% cheaper · +48.9⇄

4 models share rank 58 — the benchmark can't separate them$0.290–$3.00 /M · cheapest DeepSeek V3.2

DeepSeek V3.2DeepSeekPick GLM 5.3 Flash18% cheaper · +46.5⇄
Mistral Medium 3.5MistralPick Muse Spark 1.333% cheaper · +68.7⇄
Qwen3 VL 235B A22B InstructQwenPick GLM 5.3 Flash36% cheaper · +50.8⇄
DeepSeek V3.1 TerminusDeepSeekPick GLM 5.3 Flash48% cheaper · +53.6⇄

2 models share rank 60 — the benchmark can't separate them$0.453–$3.44 /M · cheapest DeepSeek V3.1

GPT-5.1OpenAIPick Muse Spark 1.342% cheaper · +67.6⇄
DeepSeek V3.1DeepSeekPick GLM 5.3 Flash48% cheaper · +51.3⇄

5 models share rank 65 — the benchmark can't separate them$0.205–$30.00 /M · cheapest Qwen3 235B A22B Instruct 2507

Qwen3 235B A22B Instruct 2507QwenPick DeepSeek V4.1 Flash11% cheaper · +43.4⇄
Claude Opus 4.1AnthropicPick Gemini 4 Argon73% cheaper · +114.7⇄
Qwen3.5-122B-A10BQwenPick MiMo-V2.6-Pro24% cheaper · +72.8⇄
Qwen3 Next 80B A3B InstructQwenPick GLM 5.3 Flash31% cheaper · +54.0⇄
Qwen3 235B A22B Thinking 2507QwenPick MiMo-V2.6-Pro55% cheaper · +75.3⇄

2 models share rank 66 — the benchmark can't separate them$0.563–$0.850 /M · cheapest Gemini 3.1 Flash Lite Preview

Gemini 2.5 FlashGooglePick Gemini 3.5 Flash Litesame price · +17.7⇄
Gemini 3.1 Flash Lite PreviewGooglePick MiMo-V2.6-Pro no file input4% cheaper · +74.5⇄
Inkling SmallThinking MachinesPick MiMo-V2.573% cheaper · +15.6⇄

2 models share rank 70 — the benchmark can't separate them$1.69–$4.81 /M · cheapest GPT-5.4 Mini

GPT-5.2OpenAIPick Muse Spark 1.358% cheaper · +77.6⇄
GPT-5.4 MiniOpenAIPick GPT-5.6 Luna73% cheaper · +19.0⇄

2 models share rank 74 — the benchmark can't separate them$0.536–$3.50 /M · cheapest Qwen3.5-27B

o3OpenAIPick Muse Spark 1.343% cheaper · +79.9⇄
Qwen3.5-27BQwenPick GLM 5.3 Flash56% cheaper · +62.0⇄
GPT-5OpenAIPick Muse Spark 1.342% cheaper · +83.6⇄
MiniMax M2.7MiniMaxPick GLM 5.3 Flash55% cheaper · +66.5⇄

2 models share rank 82 — the benchmark can't separate them$0.150–$3.00 /M · cheapest Step 3.5 Flash

Step 3.5 FlashStepFunPick Gemma 4 26B A4B 15% cheaper · +31.1⇄
Grok 4.7xAIPick Muse Spark 1.333% cheaper · +89.2⇄
GPT-6 SolOpenAIPick GPT-6.1 Solsame price · +50.2⇄

2 models share rank 86 — the benchmark can't separate them$0.114–$1.56 /M · cheapest Qwen3.5-Flash

Qwen3.5-FlashQwenPick DeepSeek V4 Flash 0423 no image input +11% cheaper · +35.0⇄
Grok 4.3xAIPick MiMo-V2.6-Pro shorter max output +165% cheaper · +93.5⇄

3 models share rank 87 — the benchmark can't separate them$0.200–$2.00 /M · cheapest GPT-6 Luna

Claude Haiku 4.5AnthropicPick Muse Spark 1.3same price · +93.3⇄
Qwen3.5-35B-A3BQwenPick GLM 5.3 Flash33% cheaper · +76.5⇄
GPT-6 LunaOpenAIPick DeepSeek V4.1 Flash smaller context +19% cheaper · +71.0⇄
Solar Pro 4UpstagePick GLM 5.3 Flash55% cheaper · +83.6⇄

3 models share rank 93 — the benchmark can't separate them$0.143–$0.450 /M · cheapest Qwen3 30B A3B Instruct 2507

Qwen3 30B A3B Instruct 2507QwenPick Gemma 4 26B A4B 11% cheaper · +50.2⇄
GLM 4.5 AirZ.aiPick GLM 5.3 Flash23% cheaper · +87.3⇄
GLM 4.6VZ.aiPick GLM 5.3 Flash47% cheaper · +94.7⇄
GPT-4.1OpenAIPick Gemini 3.8 Flash14% cheaper · +115.9⇄

4 models share rank 98 — the benchmark can't separate them$0.413–$0.688 /M · cheapest Qwen3 Next 80B A3B Thinking

DeepSeek V3 0324DeepSeekPick GLM 5.3 Flash46% cheaper · +96.4⇄
GPT-5 MiniOpenAIPick GPT-5.6 Luna35% cheaper · +57.4⇄
GPT-5.4 NanoOpenAIPick GPT-5.6 Luna3% cheaper · +58.5⇄
Qwen3 Next 80B A3B ThinkingQwenPick GLM 5.3 Flash42% cheaper · +102.8⇄

2 models share rank 99 — the benchmark can't separate them$0.060–$26.25 /M · cheapest gpt-oss-120b

o1OpenAIPick Gemini 4 Argon70% cheaper · +167.4⇄
gpt-oss-120bOpenAInothing better for less⇄

2 models share rank 101 — the benchmark can't separate them$0.375–$0.850 /M · cheapest Mercury 2

Nova 2 LiteAmazonPick Gemini 3.5 Flash Litesame price · +72.9⇄
Mercury 2InceptionPick GLM 5.3 Flash37% cheaper · +116.0⇄
MiniMax M2.5MiniMaxPick GLM 5.3 Flash54% cheaper · +111.8⇄
Gemma 3 27BGooglePick gpt-oss-120b +7.7, within error shorter max output +140% cheaper · +7.7, within error⇄

2 models share rank 105 — the benchmark can't separate them$0.145–$1.93 /M · cheapest GLM 4.7 Flash

o4 MiniOpenAIPick GPT-5.6 Luna77% cheaper · +77.4⇄
GLM 4.7 FlashZ.aiPick Gemma 4 26B A4B 12% cheaper · +83.3⇄
Qwen3 32BQwenPick GLM 5.3 Flash4% cheaper · +131.0⇄

2 models share rank 109 — the benchmark can't separate them$0.525–$0.850 /M · cheapest MiniMax M2

MiniMax M1MiniMaxPick MiMo-V2.6-Pro36% cheaper · +147.7⇄
MiniMax M2MiniMaxPick GLM 5.3 Flash55% cheaper · +131.0⇄

2 models share rank 110 — the benchmark can't separate them$0.388–$0.700 /M · cheapest Trinity Large Thinking

GPT-4.1 MiniOpenAIPick GPT-5.6 Luna36% cheaper · +90.2⇄
Trinity Large ThinkingArceePick GLM 5.3 Flash39% cheaper · +130.9⇄

3 models share rank 111 — the benchmark can't separate them$0.075–$1.93 /M · cheapest Gemma 3 12B

o3 Mini HighOpenAIPick GPT-5.6 Luna77% cheaper · +94.1⇄
Gemma 3 12BGooglePick gpt-oss-120b no image input20% cheaper · +31.2⇄
GLM 4.5VZ.aiPick MiMo-V2.6-Pro40% cheaper · +157.7⇄
DeepSeek V3DeepSeekPick GLM 5.3 Flash53% cheaper · +138.6⇄
Granite 4.2 8BIBMPick gpt-oss-120b shorter max output44% cheaper · +45.8⇄
GPT-5 NanoOpenAIPick Gemma 4 26B A4B smaller context +17% cheaper · +113.9⇄

2 models share rank 120 — the benchmark can't separate them$0.215–$1.93 /M · cheapest Qwen3 30B A3B

o3 MiniOpenAIPick GPT-5.6 Luna77% cheaper · +111.2⇄
Qwen3 30B A3BQwenPick DeepSeek V4.1 Flash15% cheaper · +145.7⇄

2 models share rank 124 — the benchmark can't separate them$0.043–$7.50 /M · cheapest Gemma 3 4B

GPT-4o (2024-05-13)OpenAIPick Gemini 3.8 Flash60% cheaper · +198.3⇄
Gemma 3 4BGooglenothing better for less⇄

4 models share rank 125 — the benchmark can't separate them$0.036–$4.38 /M · cheapest gpt-oss-20b

gpt-oss-20bOpenAInothing better for less⇄
GPT-4o-mini (2024-07-18)OpenAIPick GPT-6 Luna24% cheaper · +105.2⇄
GPT-4.1 NanoOpenAIPick MiMo-V2.6-Flash no file input4% cheaper · +170.5⇄
GPT-4o (2024-08-06)OpenAIPick Gemini 3.8 Flash31% cheaper · +216.2⇄
Llama 3.3 70B InstructMetaPick GLM 5.3 Flash15% cheaper · +196.8⇄

2 models share rank 130 — the benchmark can't separate them$0.370–$15.00 /M · cheapest Qwen2.5 72B Instruct

GPT-4 TurboOpenAIPick Gemini 4 Argon47% cheaper · +261.6⇄
Qwen2.5 72B InstructQwenPick GLM 5.3 Flash36% cheaper · +202.0⇄
Mistral Large 2407MistralPick Muse Spark 1.333% cheaper · +223.7⇄
Llama 3.1 70B InstructMetaPick MiMo-V2.6-Pro25% cheaper · +229.4⇄

4 models share rank 135 — the benchmark can't separate them$0.058–$4.38 /M · cheapest Mistral Small 3

Mistral Small 3MistralPick Gemma 3 4B26% cheaper · +57.3⇄
Gemma 2 27BGooglePick MiMo-V2.6-Pro17% cheaper · +259.0⇄
Qwen2.5 Coder 32B InstructQwenPick MiMo-V2.6-Pro28% cheaper · +260.3⇄
Command R+ (08-2024)CoherePick Gemini 3.8 Flash31% cheaper · +269.8⇄
Phi 4MicrosoftPick gpt-oss-120b31% cheaper · +148.8⇄

2 models share rank 140 — the benchmark can't separate them$0.025–$0.263 /M · cheapest Llama 3.1 8B Instruct

Command R (08-2024)CoherePick GLM 5.3 Flash10% cheaper · +283.7⇄
Llama 3.1 8B InstructMetanothing better for less⇄
Llama 3.2 3B InstructMetaPick DeepSeek V4 Flash 04236% cheaper · +322.7⇄
Llama 3.2 1B InstructMetaPick gpt-oss-120b15% cheaper · +310.8⇄

Just released, awaiting a score

  • Glimpse Elorian · released Oct 17, 2026 $1.20/M · 262K context
  • Moondream 3 Preview Moondream · released Oct 15, 2026 $0.850/M · 32K context
  • Moondream 3.1 Moondream · released Oct 15, 2026 $0.850/M · 32K context
  • Pareto 26.10 Unbiased · released Oct 15, 2026 $1.40/M · 1M context

212 models have no score on this board

Grouped by maker, newest first. Unscored here is not the same as scored badly.

OpenAI 35

Qwen 24

Mistral 15

Google 14

ByteDance 7

Meta 7

NVIDIA 7

Aion Labs 6

Cite this claim

Plain text and BibTeX
Plain 
131 of 143 rated, priced models have an alternative with an equal-or-higher measured score and an equal-or-lower price, with a strict advantage on at least one axis. 12 are not dominated on these two axes. As of 2026-10-21. Lens: LMArena Elo (CC BY 4.0). Ranking source digest: 5a4e1cbffc4e5f72. https://undominated.ai/
Publication: sha256:7bd9c5d7a89db63c6f9d85d4cca8a8c5232d9f5020cfec3282bcc212f903f16d; calculation SHA-256: 7cb91c0d61fb42b1acdcbd4eba37b97c07f5debd871503fc334a3ace79f82bbf; permitted inputs: https://undominated.ai/data/publications/manifests/7bd9c5d7a89db63c6f9d85d4cca8a8c5232d9f5020cfec3282bcc212f903f16d.json
BibTeX 
@misc{undominated-dominated-census-2026-10-21,
  author = {{Undominated.ai}},
  title  = {Dominated census — Undominated.ai},
  year   = {2026},
  url    = {https://undominated.ai/},
  note   = {as of 2026-10-21; LMArena Elo (CC BY 4.0); ranking source 5a4e1cbffc4e5f72},
  annote = {Publication: sha256:7bd9c5d7a89db63c6f9d85d4cca8a8c5232d9f5020cfec3282bcc212f903f16d; calculation SHA-256: 7cb91c0d61fb42b1acdcbd4eba37b97c07f5debd871503fc334a3ace79f82bbf; permitted inputs: https://undominated.ai/data/publications/manifests/7bd9c5d7a89db63c6f9d85d4cca8a8c5232d9f5020cfec3282bcc212f903f16d.json}
}

LMArena scores used under CC BY 4.0 · numbers recomputed from the catalogue on this page. Input identity & calculation SHA.

Evidence & Ask
Markdown for LLMs