45 modelschecked Jun 25, 2026, 11:59 PM UTCdaily88 failed refresheshow we measure this →CSV / JSON

Follow changes

As of

Cheapest today

DeepSeek V4 Flash

$0.016LiveBench cost per successful task

Dearest published row: $1.573
Middle of the board: $0.24

Highest $1.573Middle $0.24

Rank 1one bar per published rowRank 45

Official LiveBench release: 2026-06-25 · 7 checks behind each row · 100% confidence across rows.

LiveBench model cost efficiency current rankings
Rank	Name	LiveBench cost per successful task	Livebench average input tokens	Livebench average output tokens	Livebench cost per question	Why
01	DeepSeek V4 Flash	$0.016	148.8K	34.4K	$0.01	This rank
02	Grok Build 0.1	$0.024	58.8K	980	$0.02	This rank
03	DeepSeek V4 Pro	$0.050	96.6K	35K	$0.04	This rank
04	MiniMax M3	$0.060	331.9K	15.5K	$0.04	This rank
05	Grok 4.3	$0.061	28.7K	11.2K	$0.04	This rank
06	Gemini 3.5 Flash Lite High	$0.069	112.1K	11.5K	$0.04	This rank
07	GPT 5.4 Nano xHigh	$0.091	180.3K	46.2K	$0.06	This rank
08	Kimi K2.7 Code	$0.100	138.1K	13K	$0.07	This rank
09	GPT 5.6 Luna xHigh	$0.123	45.3K	13.1K	$0.09	This rank
10	Grok 4.5	$0.128	61.2K	10.5K	$0.1	This rank
11	Muse Spark 1.1 High	$0.138	296.7K	13.8K	$0.1	This rank
12	Kimi K2.6 Thinking	$0.169	78K	27K	$0.12	This rank
13	Qwen3.7 Max	$0.182	60.7K	12.9K	$0.13	This rank
14	GPT 5.2 Codex	$0.187	37K	9,154	$0.14	This rank
15	GLM 5.2	$0.195	229.4K	23.5K	$0.14	This rank
16	Qwen3.6 27b	$0.202	69.3K	24.3K	$0.13	This rank
17	GPT 5.6 Luna Max	$0.202	126K	21.8K	$0.15	This rank
18	GPT 5.6 Terra xHigh	$0.208	77.4K	8,228	$0.15	This rank
19	GPT 5.5	$0.212	23.7K	4,151	$0.15	This rank
20	Qwen3.6 Plus	$0.227	143.9K	28.2K	$0.16	This rank
21	GPT 5.2 2025 12 11 High	$0.234	50.8K	11.5K	$0.17	This rank
22	Muse Spark 1.1 xHigh	$0.234	200.1K	13.7K	$0.18	This rank
23	Gemini 3.6 Flash High	$0.235	188.1K	14.7K	$0.17	This rank
24	Gemini 3.5 Flash High	$0.249	108.3K	15.5K	$0.19	This rank
25	Gemini 3.1 Pro Preview High	$0.262	99.1K	13.4K	$0.2	This rank
26	Claude Sonnet 4 6 Thinking Auto Medium Effort	$0.306	1,328	13.2K	$0.22	This rank
27	GPT 5.4 Mini xHigh	$0.334	70.7K	47.4K	$0.22	This rank
28	Inkling xHigh	$0.343	355.6K	22.2K	$0.25	This rank
29	GPT 5.6 Sol xHigh	$0.355	86.8K	7,161	$0.28	This rank
30	GPT 5.5 High	$0.367	62.6K	7,718	$0.29	This rank
31	Claude Opus 5 High Effort	$0.373	1,265	8,374	$0.29	This rank
32	Kimi K3	$0.379	261.8K	13.6K	$0.3	This rank
33	GPT 5.4 xHigh	$0.387	77K	18.3K	$0.3	This rank
34	Claude Opus 4 6 Thinking Auto High Effort	$0.404	1,935	9,651	$0.3	This rank
35	Claude Opus 5 xHigh Effort	$0.487	1,248	11.1K	$0.39	This rank
36	Claude Sonnet 5 xHigh Effort	$0.492	1,214	19.4K	$0.37	This rank
37	GPT 5.6 Terra Max	$0.497	179.1K	22.1K	$0.4	This rank
38	Claude Opus 4 7 xHigh Effort	$0.528	1,285	11.5K	$0.4	This rank
39	GPT 5.5 xHigh	$0.530	95.9K	11.4K	$0.42	This rank
40	GPT 5.6 Sol Max	$0.589	181.2K	11.7K	$0.49	This rank
41	Claude Opus 4 5 20251101 Thinking 64K High Effort	$0.610	79K	15.2K	$0.44	This rank
42	Claude Opus 5 Max Effort	$0.641	1,263	14.8K	$0.51	Showing
43	Claude Opus 4 8 xHigh Effort	$0.688	1,259	17.1K	$0.54	This rank
44	Claude Fable 5 xHigh Effort	$0.811	1,347	10.5K	$0.64	This rank
45	Claude Fable 5 Max Effort	$1.573	1,264	20.3K	$1.27	This rank

Some explanation inputs are unavailable Missing: publicFrozenEvidence, comparablePriorRanking.

Published rank: #42
Score: 0.641
Sample size: 7

100% confidence · 64% component coverage · as of Jun 25, 2026, 11:59 PM UTC

SCORE CONSTRUCTION

Component ledger

7/11 evidenced

Components contributing to Claude Opus 5 Max Effort's rank
Component	Value	Weight	Contribution	Evidence	Status
livebench average input tokens	—	0%	0	1	available
livebench average output tokens	—	0%	0	1	available
livebench cost per question USD	—	0%	0	1	available
livebench cost per successful task USD	—	100%	0.641	1	available
livebench input price per million USD	—	0%	0	1	available
livebench output price per million USD	—	0%	0	1	available
livebench overall score	—	0%	0	1	available
metrics	Structured	Unavailable	—	0	unavailableno_public_frozen_evidence
rankingDirection	ascending	Unavailable	—	0	unavailableno_public_frozen_evidence
release	2026-06-25	Unavailable	—	0	unavailableno_public_frozen_evidence
sourceModelId	claude-opus-5-max-effort	Unavailable	—	0	unavailableno_public_frozen_evidence

WHY IT MOVED

new since prior snapshot

No rank delta

no_prior_published_snapshot

No component moved since the prior snapshot.

PRIMARY EVIDENCE

Source trail

7 public records

LiveBench average input tokenslivebench average input tokens · observed Jun 25, 2026, 11:59 PM UTC1,263 tokens100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench average output tokenslivebench average output tokens · observed Jun 25, 2026, 11:59 PM UTC14,826 tokens100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench cost per questionlivebench cost per question USD · observed Jun 25, 2026, 11:59 PM UTC$0.51100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench cost per successful tasklivebench cost per successful task USD · observed Jun 25, 2026, 11:59 PM UTC$0.64100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench input pricelivebench input price per million USD · observed Jun 25, 2026, 11:59 PM UTC$5.00100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench output pricelivebench output price per million USD · observed Jun 25, 2026, 11:59 PM UTC$25.00100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗
LiveBench overall scorelivebench overall score · observed Jun 25, 2026, 11:59 PM UTC79.234 percent100% confidence
Source
LiveBench official model benchmark
Grade
A
Freshness
fresh
Locator
{"formula":"(sum task cost / sum question count) / (overall score / 100)","costAsset":"https://livebench.ai/cost_2026_06_25.csv","scoreAsset":"https://livebench.ai/table_2026_06_25.csv","categoryAsset":"https://livebench.ai/categories_2026_06_25.json"}
Open primary evidence ↗

Embed this table

Paste this on your site. The table stays current automatically and the credit line links readers to the full ranking.

<iframe src="https://www.rank.ai/embed/boards/livebench-model-efficiency" title="LiveBench Model Cost Efficiency" width="100%" height="420" loading="lazy" style="border:1px solid #111;max-width:720px"></iframe>
<p style="font-size:12px">Source: <a href="https://www.rank.ai/rankings/efficiency/livebench-model-efficiency">LiveBench Model Cost Efficiency</a> by Rank.ai</p>

More in Model efficiency

Capability relative to token price, context, latency, and throughput.

3 live tables

Model efficiency

Licensed capability-per-dollar rankings with exact model cohort, price basis, evidence, and freshness gates.

Quality per dollar

Intelligence points per weighted inference dollar at public API prices.

Context per dollar

Usable context capacity relative to published input-token pricing.

Model efficiency

Licensed capability-per-dollar rankings with exact model cohort, price basis, evidence, and freshness gates.

Quality per dollar

Intelligence points per weighted inference dollar at public API prices.

Context per dollar

Usable context capacity relative to published input-token pricing.

Get 10 Free
AI-Generated Articles

LiveBench model cost efficiency

Claude Opus 5 Max Effort

Component ledger

new since prior snapshot

Source trail

More in Model efficiency

Local SEO

AI Visibility

Content

Tools & Free

Compare

Resources

Company