What is the most used AI model for coding?

Measured by real token burn from developers' machines on whoburnedmore, the top-ranked model on this page currently leads real coding usage. Unlike benchmark leaderboards, this ranking reflects what developers actually run in Claude Code, Codex, Cursor, Gemini CLI and other coding agents.

How is this different from OpenRouter's model rankings?

OpenRouter ranks models by API traffic routed through its gateway. This page measures usage that never touches an API gateway — subscription products like Claude Code Max plans, Cursor and Codex — read directly from the local usage logs those tools keep, via the open-source whoburnedmore CLI.

Where does this data come from?

From developers who run npx whoburnedmore and choose to be listed publicly. The CLI reads each coding agent's own local logs and submits daily aggregate token totals per tool and model — never prompts, code, or file names.

WHOBURNEDMORE// LIVE USAGE TERMINAL

LIVE · 2026-08-14 16:00Z

AI MODEL USAGE RANKINGS — real coding burn, last 30 days

measured from local agent logs · includes subscription usage gateways never see

Tokens measured

748.59B

last 30 days

Estimated cost

$471,871

Developers

144

listed panel

Models

Cache hit

97%

prompt-side

Burn / dev

5.20B

#	Model	Devs	Tokens	Share	Input	Output	Cache	Est cost
01	gpt-5.6-solCodex CLI · OpenCode · Hermes +1	64	410.10B	55%	11.16B	1.12B	97%	$202,259
02	claude-opus-5Claude Code · Hermes · OpenCode	64	80.11B	11%	211.8M	355.5M	100%	$60,837
03	claude-opus-4-8Claude Code · Hermes · OpenCode	74	78.94B	11%	618.2M	565.6M	99%	$72,030
04	claude-fable-5Claude Code · OpenCode	64	39.81B	5.3%	509.5M	454.3M	99%	$85,639
05	claude-sonnet-5Claude Code · Hermes	62	37.42B	5.0%	18.9M	149.4M	100%	$11,272
06	gpt-5.5Codex CLI · OpenCode · Hermes	54	14.44B	1.9%	719.6M	39.0M	95%	$10,225
07	gpt-5.6-terraCodex CLI · Hermes · Claude Code +1	68	13.70B	1.8%	483.3M	50.4M	96%	$5,479
08	gpt-5.6-lunaCodex CLI · OpenCode · Claude Code	37	11.14B	1.5%	328.7M	31.9M	97%	$1,086
09	gpt-5Codex CLI	12	9.71B	1.3%	356.5M	19.0M	96%	$846
10	codex-auto-reviewCodex CLI	27	7.89B	1.1%	353.8M	20.9M	96%	$695
11	deepseek-v4-flash-freeOpenCode · Hermes · Claude Code	27	6.51B	0.9%	234.6M	30.5M	96%	$0.06
12	defaultCursor · OpenCode	17	5.93B	0.8%	373.1M	33.1M	94%	$3,761
13	deepseek-v4-flashOpenCode · Kimi CLI · Claude Code +1	15	4.08B	0.5%	117.7M	10.3M	97%	$38.49
14	cursor-grok-4.5-high-fastCursor	18	3.86B	0.5%	352.7M	30.6M	91%	$5,337
15	claude-opus-4-6Claude Code	7	2.63B	0.4%	342.7K	16.5M	100%	$2,077
16	claude-sonnet-4-6Claude Code	22	1.91B	0.3%	2.0M	11.5M	100%	$976
17	k3Kimi CLI · Claude Code · OpenCode	4	1.83B	0.2%	60.7M	9.6M	97%	$840
18	rollupClaude Code	3	1.74B	0.2%	1.74B	0	0.0%	$1,899
19	claude-haiku-4-5-20251001Claude Code	47	1.69B	0.2%	421.9K	13.8M	100%	$358
20	gemini-3.6-flashOpenCode	1	1.05B	0.1%	75.6M	1.2M	93%	$278
21	glm-5.2OpenCode · Hermes · Codex CLI +1	12	978.3M	0.1%	199.8M	5.4M	79%	$461
22	cursor-grok-4.5-highCursor	9	976.2M	0.1%	73.9M	7.0M	92%	$563
23	unreportedHermes · OpenCode · Factory Droid +1	11	956.3M	0.1%	165.8M	4.7M	83%	$34.82
24	deepseek-v4-proClaude Code · OpenCode · Hermes	10	899.2M	0.1%	18.5M	1.5M	98%	$17.47
25	claude-opus-4-7Claude Code	4	834.6M	0.1%	3.3K	1.5M	100%	$567
26	composer-2.5-fastCursor	23	825.7M	0.1%	52.9M	7.0M	94%	$643
27	gpt-5.4Codex CLI · Hermes	14	686.0M	<0.1%	32.1M	3.3M	95%	$370
28	gpt-daybreak-blue-latestCodex CLI	1	557.2M	<0.1%	32.0M	2.3M	94%	$396
29	gpt-5.3-codex-sparkCodex CLI	7	513.7M	<0.1%	25.3M	2.2M	95%	$146
30	glm-5.2-highCursor	4	495.3M	<0.1%	7.0M	1.6M	99%	$141
31	claude-4.6-sonnet-mediumCursor	1	480.6M	<0.1%	7.5M	2.1M	98%	$235
32	9router_freeHermes	1	387.0M	<0.1%	86.9M	1.1M	77%	$0
33	stealthClaude Code	1	353.7M	<0.1%	28.6M	1.0M	92%	$0
34	minimax-m3OpenCode · Hermes · Claude Code	5	299.1M	<0.1%	13.9M	1.7M	95%	$16.83
35	step-3.7-flash:freeClaude Code · Hermes · Codex CLI	2	298.0M	<0.1%	258.8M	2.1M	13%	$0
36	claude-sonnet-5-thinking-highCursor	6	295.3M	<0.1%	782.7K	1.4M	100%	$115
37	gpt-5.6-sol-highCursor	2	269.9M	<0.1%	4.0M	1.3M	98%	$234
38	claude-opus-5-thinking-highCursor	7	263.2M	<0.1%	917.2K	1.8M	100%	$248
39	claude-opus-4-5OpenCode	1	235.5M	<0.1%	21.8K	1.8M	100%	$219
40	big-pickleOpenCode · Claude Code	11	224.7M	<0.1%	41.0M	1.1M	82%	$0
41	kimi-k3OpenCode · Claude Code · Codex CLI	10	192.6M	<0.1%	12.5M	928.7K	93%	$118
42	gemini-3.6-flash-highCursor	2	181.3M	<0.1%	34.2M	680.1K	81%	$68.66
43	nemotron-3-ultra-freeOpenCode · Claude Code	5	172.7M	<0.1%	35.4M	330.7K	79%	$0
44	gpt-5.3-codexCursor	2	149.3M	<0.1%	18.2M	838.0K	88%	$65.74
45	claude-fable-5-thinking-highCursor	6	137.4M	<0.1%	1.3M	1.1M	99%	$289
46	cursor-grok-4.6-high-fastCursor	6	122.7M	<0.1%	8.2M	981.7K	93%	$158
47	hy3:freeHermes	3	120.5M	<0.1%	3.7M	1.8M	97%	$0
48	composer-2.5Cursor	5	115.8M	<0.1%	8.3M	690.5K	93%	$26.91
49	model-237bf97a0500Pi	2	112.4M	<0.1%	8.6M	410.6K	92%	$5.75
50	kimi-for-codingClaude Code · OpenCode · Kimi CLI	4	109.9M	<0.1%	5.8M	515.9K	95%	$14.17
51	autoOpenCode	1	104.5M	<0.1%	6.6M	115.3K	94%	$0
52	model-ba4f0d6a6174Pi	1	90.2M	<0.1%	1.1M	451.8K	99%	$0.5
53	[pi] deepseek-v4-flashPi	1	84.8M	<0.1%	961.3K	388.7K	99%	$0.47
54	kimi-for-coding-highspeedKimi CLI · Claude Code	2	84.0M	<0.1%	1.9M	321.3K	98%	$0.02
55	kimi-k2.7-codeOpenCode	2	77.6M	<0.1%	6.0M	389.4K	92%	$21.19
56	gpt-5.4-miniCodex CLI · Cursor	10	76.1M	<0.1%	4.7M	512.9K	94%	$19.98
57	claude-opus-4-8-thinking-highCursor	5	67.2M	<0.1%	330.9K	373.0K	99%	$62.19
58	nemotron-3-super-120b-a12bClaude Code	1	63.1M	<0.1%	62.8M	345.8K	0.0%	$0
59	mimo-v2.5-freeClaude Code · OpenCode · Hermes	6	58.5M	<0.1%	39.5M	136.4K	32%	$15.44
60	cursor-grok-4.5-mediumCursor	2	56.5M	<0.1%	7.3M	503.3K	87%	$41.95

Model

Devs

Tokens

Input

Output

Cache

Est cost

gpt-5.6-solCodex CLI · OpenCode · Hermes +1

410.10B

55%

11.16B

1.12B

97%

$202,259

claude-opus-5Claude Code · Hermes · OpenCode

80.11B

11%

211.8M

355.5M

100%

$60,837

claude-opus-4-8Claude Code · Hermes · OpenCode

78.94B

11%

618.2M

565.6M

99%

$72,030

claude-fable-5Claude Code · OpenCode

39.81B

5.3%

509.5M

454.3M

99%

$85,639

claude-sonnet-5Claude Code · Hermes

37.42B

5.0%

18.9M

149.4M

100%

$11,272

gpt-5.5Codex CLI · OpenCode · Hermes

14.44B

1.9%

719.6M

39.0M

95%

$10,225

gpt-5.6-terraCodex CLI · Hermes · Claude Code +1

13.70B

1.8%

483.3M

50.4M

96%

$5,479

gpt-5.6-lunaCodex CLI · OpenCode · Claude Code

11.14B

1.5%

328.7M

31.9M

97%

$1,086

gpt-5Codex CLI

9.71B

1.3%

356.5M

19.0M

96%

$846

codex-auto-reviewCodex CLI

7.89B

1.1%

353.8M

20.9M

96%

$695

deepseek-v4-flash-freeOpenCode · Hermes · Claude Code

6.51B

0.9%

234.6M

30.5M

96%

$0.06

defaultCursor · OpenCode

5.93B

0.8%

373.1M

33.1M

94%

$3,761

deepseek-v4-flashOpenCode · Kimi CLI · Claude Code +1

4.08B

0.5%

117.7M

10.3M

97%

$38.49

cursor-grok-4.5-high-fastCursor

3.86B

0.5%

352.7M

30.6M

91%

$5,337

claude-opus-4-6Claude Code

2.63B

0.4%

342.7K

16.5M

100%

$2,077

claude-sonnet-4-6Claude Code

1.91B

0.3%

2.0M

11.5M

100%

$976

k3Kimi CLI · Claude Code · OpenCode

1.83B

0.2%

60.7M

9.6M

97%

$840

rollupClaude Code

1.74B

0.2%

1.74B

0.0%

$1,899

claude-haiku-4-5-20251001Claude Code

1.69B

0.2%

421.9K

13.8M

100%

$358

gemini-3.6-flashOpenCode

1.05B

0.1%

75.6M

1.2M

93%

$278

glm-5.2OpenCode · Hermes · Codex CLI +1

978.3M

0.1%

199.8M

5.4M

79%

$461

cursor-grok-4.5-highCursor

976.2M

0.1%

73.9M

7.0M

92%

$563

unreportedHermes · OpenCode · Factory Droid +1

956.3M

0.1%

165.8M

4.7M

83%

$34.82

deepseek-v4-proClaude Code · OpenCode · Hermes

899.2M

0.1%

18.5M

1.5M

98%

$17.47

claude-opus-4-7Claude Code

834.6M

0.1%

3.3K

1.5M

100%

$567

composer-2.5-fastCursor

825.7M

0.1%

52.9M

7.0M

94%

$643

gpt-5.4Codex CLI · Hermes

686.0M

<0.1%

32.1M

3.3M

95%

$370

gpt-daybreak-blue-latestCodex CLI

557.2M

<0.1%

32.0M

2.3M

94%

$396

gpt-5.3-codex-sparkCodex CLI

513.7M

<0.1%

25.3M

2.2M

95%

$146

glm-5.2-highCursor

495.3M

<0.1%

7.0M

1.6M

99%

$141

claude-4.6-sonnet-mediumCursor

480.6M

<0.1%

7.5M

2.1M

98%

$235

9router_freeHermes

387.0M

<0.1%

86.9M

1.1M

77%

stealthClaude Code

353.7M

<0.1%

28.6M

1.0M

92%

minimax-m3OpenCode · Hermes · Claude Code

299.1M

<0.1%

13.9M

1.7M

95%

$16.83

step-3.7-flash:freeClaude Code · Hermes · Codex CLI

298.0M

<0.1%

258.8M

2.1M

13%

claude-sonnet-5-thinking-highCursor

295.3M

<0.1%

782.7K

1.4M

100%

$115

gpt-5.6-sol-highCursor

269.9M

<0.1%

4.0M

1.3M

98%

$234

claude-opus-5-thinking-highCursor

263.2M

<0.1%

917.2K

1.8M

100%

$248

claude-opus-4-5OpenCode

235.5M

<0.1%

21.8K

1.8M

100%

$219

big-pickleOpenCode · Claude Code

224.7M

<0.1%

41.0M

1.1M

82%

kimi-k3OpenCode · Claude Code · Codex CLI

192.6M

<0.1%

12.5M

928.7K

93%

$118

gemini-3.6-flash-highCursor

181.3M

<0.1%

34.2M

680.1K

81%

$68.66

nemotron-3-ultra-freeOpenCode · Claude Code

172.7M

<0.1%

35.4M

330.7K

79%

gpt-5.3-codexCursor

149.3M

<0.1%

18.2M

838.0K

88%

$65.74

claude-fable-5-thinking-highCursor

137.4M

<0.1%

1.3M

1.1M

99%

$289

cursor-grok-4.6-high-fastCursor

122.7M

<0.1%

8.2M

981.7K

93%

$158

hy3:freeHermes

120.5M

<0.1%

3.7M

1.8M

97%

composer-2.5Cursor

115.8M

<0.1%

8.3M

690.5K

93%

$26.91

model-237bf97a0500Pi

112.4M

<0.1%

8.6M

410.6K

92%

$5.75

kimi-for-codingClaude Code · OpenCode · Kimi CLI

109.9M

<0.1%

5.8M

515.9K

95%

$14.17

autoOpenCode

104.5M

<0.1%

6.6M

115.3K

94%

model-ba4f0d6a6174Pi

90.2M

<0.1%

1.1M

451.8K

99%

$0.5

[pi] deepseek-v4-flashPi

84.8M

<0.1%

961.3K

388.7K

99%

$0.47

kimi-for-coding-highspeedKimi CLI · Claude Code

84.0M

<0.1%

1.9M

321.3K

98%

$0.02

kimi-k2.7-codeOpenCode

77.6M

<0.1%

6.0M

389.4K

92%

$21.19

gpt-5.4-miniCodex CLI · Cursor

76.1M

<0.1%

4.7M

512.9K

94%

$19.98

claude-opus-4-8-thinking-highCursor

67.2M

<0.1%

330.9K

373.0K

99%

$62.19

nemotron-3-super-120b-a12bClaude Code

63.1M

<0.1%

62.8M

345.8K

0.0%

mimo-v2.5-freeClaude Code · OpenCode · Hermes

58.5M

<0.1%

39.5M

136.4K

32%

$15.44

cursor-grok-4.5-mediumCursor

56.5M

<0.1%

7.3M

503.3K

87%

$41.95

AI MODEL USAGE RANKINGS — real coding burn, last 30 days

AI MODEL USAGE RANKINGS — real coding burn, last 30 days

Models by real token burn

Daily burn × model

AI MODEL USAGE RANKINGS — real coding burn, last 30 days

AI MODEL USAGE RANKINGS — real coding burn, last 30 days

▮Models by real token burn

▮Daily burn × model

Models by real token burn

Daily burn × model