Live · trailing 30 days

Model leaderboard

The models developers actually run on boldrouter, ranked by real traffic across every provider - popularity, speed, and reliability, updated from live usage.

Requests

17.6K

Tokens

55.8M

Models in use

Providers

Most popularClaude Sonnet 5Anthropic5,816 reqMost requests

FastestText Embedding 3 SmallOpenAI513msLowest avg latency

Best valueText Embedding 3 SmallOpenAI$0.02Cheapest blended price

Most reliableClaude Sonnet 5Anthropic100.0%Highest success rate

Models by usage

Ranked by request volume. Share is each model's slice of all platform traffic over the window.

#	Model	Requests	Share	Tokens	Avg latency	Success	From / 1M
1	Claude Sonnet 5 Anthropic	5,816	33.0%	24.7M	969ms	100.0%	$12.00
2	Claude Haiku 4.5 Anthropic	5,668	32.2%	9.5M	749ms	100.0%	$6.00
3	GPT-4o mini OpenAI	1,849	10.5%	13.4M	3109ms	100.0%	$0.75
4	Gemini 2.5 Flash Google	949	5.4%	1.2M	5782ms	96.9%	$2.80
5	Sonar Pro Perplexity	860	4.9%	751.5K	3771ms	100.0%	$18.00
6	GPT-4.1 mini OpenAI	858	4.9%	512.9K	2159ms	100.0%	$2.00
7	Grok 4.3 xAI	858	4.9%	591.6K	11676ms	100.0%	$3.75
8	Claude Opus 5.5 Anthropic	146	0.8%	2.5M	8201ms	93.2%	$24.00
9	Bold Frontierauto · 4 Bold	129	0.7%	872.1K	4751ms	100.0%	$3.75
10	Gemini 3.1 Flash Image (Nano Banana 2) Google	113	0.6%	182K	9833ms	98.2%	$3.50
11	Gemini 3 Pro Image (Nano Banana Pro) Google	75	0.4%	115.8K	18997ms	98.7%	$14.00
12	Gemini 2.5 Flash-Lite Google	68	0.4%	716.5K	3411ms	100.0%	$0.50
13	Text Embedding 3 Small OpenAI	65	0.4%	1,103	513ms	100.0%	$0.02
14	GPT-OSS 120B W&B	47	0.3%	321.7K	31837ms	97.9%	$0.20
15	Bold Autoauto · 5 Bold	27	0.2%	27K	2230ms	100.0%	$1.50
16	GPT Image 2 OpenAI	12	0.1%	4,404	16343ms	100.0%	$5.00
17	GPT-4o OpenAI	11	0.1%	55K	5305ms	100.0%	$12.50
18	DeepSeek V4 Flash DeepSeek	5	0.0%	907	858ms	100.0%	$1.50
19	FLUX.2 Max FLUX	1	0.0%	117	25994ms	100.0%	-
20	FLUX.2 Pro FLUX	1	0.0%	117	11159ms	100.0%	-
21	Bold Imageauto · 5 Bold	1	0.0%	117	6116ms	100.0%	-
22	Gemini 2.5 Flash Image Google	1	0.0%	1,364	7884ms	100.0%	$2.80
23	GPT Image 1 OpenAI	1	0.0%	6,317	45933ms	100.0%	$5.00
24	Grok Imagine Image xAI	1	0.0%	117	8076ms	100.0%	-

Providers by usage

The same traffic rolled up by vendor. Latency and success are request-weighted across each vendor's models.

#1Anthropic3 models

66%

66.0%

Requests

11.6K

Avg latency

953ms

Success

99.9%

#2OpenAI6 models

16%

15.9%

Requests

2,796

Avg latency

2838ms

Success

100.0%

#3Google5 models

6.8%

Requests

1,206

Avg latency

6851ms

Success

97.3%

#4Perplexity1 models

4.9%

Requests

860

Avg latency

3771ms

Success

100.0%

#5xAI2 models

4.9%

Requests

859

Avg latency

11672ms

Success

100.0%

#6Bold4 models

1.0%

Requests

173

Avg latency

3984ms

Success

96.0%

#7W&B1 models

0.3%

Requests

Avg latency

31837ms

Success

97.9%

#8Z.ai1 models

0.3%

Requests

Avg latency

50179ms

Success

87.2%

#9DeepSeek1 models

0.0%

Requests

Avg latency

858ms

Success

100.0%

#10FLUX2 models

0.0%

Requests

Avg latency

18577ms

Success

100.0%

How this is measured

Figures aggregate metered requests across all Bold Router accounts over the trailing 30 days, grouped by model. Only model-level totals are shown - never per-account usage, spend, or identity. In-dashboard playground traffic is excluded. Prices are each model's cheapest provider route at list price (input + output per 1M tokens). “Best value” and “Fastest” awards require at least 5 requests in the window.