Best LLM APIs by intelligence per dollar

A second take on value, from a different benchmark. This ranks56 models by the Artificial Analysis Intelligence Index — an aggregate 0–100 intelligence score — divided by each model's cheapest blended price per 1M tokens. Where the Elo value boardmeasures human preference, this measures aggregated reasoning and knowledge. Each model links to its cheapest provider; all figures computed from the index.

Ranked by intelligence per dollar

#ModelIndexCheapest /1MVerifiedIndex / $
1deepseek-v4-flash · Baidu51.8$0.15today345.3
2glm-5.3-flash · Relace57.5$0.3091d ago186.2
3nemotron-3.5-lightning · DeepInfra23.6$0.28today84.3
4granite-4.2-8b · CoreWeave19.6$0.25today78.4
5gpt-5.6-luna · OpenAI51.2$0.7today73.1
6gemma-4-31b · CoreWeave29.7$0.44today67.5
7gemma-4-26b-a4b · Cloudflare26.1$0.4today65.2
8nemotron-3-super-120b-a12b · DeepInfra25.7$0.485today53
9glm-4.7-flash · Venice23.3$0.46today50.7
10deepseek-v3.2 · GMICloud25.1$0.518today48.4
11minimax-m3 · CoreWeave45.4$1.19today38.2
12ministral-3b · Mistral AI7.1$0.2today35.5
13qwen3.6-35b-a3b · AkashML32.1$1today32.1
14ministral-8b · Mistral AI9$0.3today30
15minimax-m2.7 · Novita AI38.9$1.35today28.8
16minimax-m2.5 · Venice34.5$1.22today28.3
17ministral-14b · Mistral AI11.2$0.4today28
18qwen3.5-35b-a3b · DeepInfra29.9$1.14today26.2
19qwen3.8-27b · Parasail52$2.44today21.3
20qwen3.5-27b · Alibaba34.6$1.755today19.7
21qwen3.6-27b · Chutes37.7$2.3today16.4
22glm-5 · GMICloud40.5$2.52today16.1
23glm-4.7 · DeepInfra34.5$2.15today16
24kimi-k2.6 · Baidu45.1$2.856today15.8
25nemotron-3-ultra-550b-a55b · DeepInfra38.3$2.7today14.2
26qwen3.5-122b-a10b · SiliconFlow32.9$2.34today14.1
27kimi-k2.5 · SiliconFlow36$2.7today13.3
28gemini-3.8-flash · Google58.7$4.52d ago13
29glm-5.3 · Reka59.5$4.65today12.8
30qwen3.5-397b-a17b · Alibaba34.3$2.73today12.6
31gemini-3.6-flash · Google51.6$4.52d ago11.5
32glm-5.1 · Baidu41$3.764today10.9
33glm-4.6 · Venice23.4$2.18today10.7
34kimi-k2.7-code · DeepInfra43$4.08today10.5
35gpt-5.6-sol · OpenAI58.9$6today9.8
36gpt-5.6-terra · OpenAI55$7today7.9
37grok-4.6 · xAI60.9$82d ago7.6
38claude-sonnet-5 · Anthropic55.3$12today4.6
39kimi-k3 · DeepInfra59.7$17.1today3.5
40mistral-medium-3-5 · Mistral AI30.4$9today3.4
41claude-opus-5 · Anthropic63$30today2.1
42claude-fable-5 · Anthropic62.1$60today1

Value = index score ÷ (input + output price per 1M tokens). A model's score is identical across providers, so the cheapest host gives the best value — that's the price shown. Models released within the last 12 months; older ones below. Blended is a coarse ranking aid, not a workload estimate; size your own on each model's page.

Older models — released more than 12 months ago (14)

Still priced daily and often still the pragmatic choice, but no longer the current generation.

#ModelIndexCheapest /1MAgeIndex / $
43gpt-oss-20b · AkashML15.2$0.1213 mo126.7
44gpt-oss-120b · AkashML24.1$0.213 mo120.5
45gemma-3-27b · DeepInfra7.4$0.2417 mo30.8
46llama-4-scout · DeepInfra10.3$0.417 mo25.8
47deepseek-chat-v3.1 · DeepInfra21.4$1.212 mo17.8
48glm-4.5-air · Novita AI16.7$0.9813 mo17
49llama-4-maverick · DigitalOcean14.5$0.89617 mo16.2
50deepseek-v3 · DeepInfra14.2$1.2120 mo11.7
51gpt-4o-mini · OpenAI6.7$0.7525 mo8.9
52deepseek-r1 · DeepInfra20.4$2.6515 mo7.7
53gpt-4.1-mini · OpenAI14.8$216 mo7.4
54minimax-m1 · Minimax17.9$2.614 mo6.9
55mistral-medium-3 · Mistral AI12.5$2.415 mo5.2
56gpt-4o · OpenAI11.1$12.527 mo0.9

For contrast: the top of the intelligence index

The highest-scoring models by raw index — rarely the best value, which is the point.

Index rankModelIndexCheapest /1MIndex / $
1claude-opus-563$302.1
2claude-fable-562.1$601
3grok-4.660.9$87.6
4kimi-k359.7$17.13.5
5glm-5.359.5$4.6512.8
6gpt-5.6-sol58.9$69.8
7gemini-3.8-flash58.7$4.513
8glm-5.3-flash57.5$0.309186.2
9claude-sonnet-555.3$124.6
10gpt-5.6-terra55$77.9

Artificial Analysis Intelligence Index · verified 2026-09-05 · leaderboard source · full methodology. Two benchmarks disagree by design — cross-check the Elo value ranking.

FAQ

Which LLM API gives the most intelligence per dollar?

As of 2026-09-05, deepseek-v4-flash leads: 345.3 index points per dollar — a 51.8 Artificial Analysis Intelligence Index score at $0.15 per 1M tokens (blended input + output) via Baidu.

How is this different from the Elo value ranking?

This ranks by the Artificial Analysis Intelligence Index — an aggregate 0–100 reasoning and knowledge score — instead of LMArena Elo, which measures human preference. Two independent benchmarks cross-check each other; they disagree by design, so a model strong on one may rank differently on the other.

Is the smartest model the best value?

Rarely. The top model by raw index is claude-opus-5 at 63, but at $30 per 1M its value is 2.1 points per dollar — below the value leader deepseek-v4-flash. Peak intelligence usually carries a price premium.