Updated Sep 1, 4:58 AM GMT+2

Model Analytics

Usage of models across Spettro. See which models are winning real usage, how the mix shifts over time, and where momentum is moving each week.

SuperFast (DeepSeek V4 Flash) leads with 45.3% of token usage, ahead of GLM 5.2 at 17.1% and MiniMax M3 at 11.1%.

Total Tokens 29.4B Jun 2026 to Sep 2026

Top Model 24h MiniMax M3 15K tokens · 100%

Top models.

Usage of models across Spettro.

SuperFast (DeepSeek V4 Flash) DeepSeek

13.3B

-27%

GLM 5.2 Zhipu

-33%

MiniMax M3 MiniMax

3.3B

-29%

Kimi K3 Moonshot

2.9B

+75%

SuperSmart (DeepSeek V4 Pro) DeepSeek

2.8B

-6%

Kimi K2.7 Coder Moonshot

1.4B

-72%

Qwen3.8 Max Qwen

538.9M

-28%

Qwen3.7 Max Qwen

118.9M

New

Qwen3.6 Flash Qwen

39M

New

Qwen3.7 Plus Qwen

5.9M

-94%

GLM 5.1 Zhipu

67.3K

New

Leaderboard.

Full ranking by token usage.

#		Model	Provider	Tokens	Market Share	Δ 7d
01	=	SuperFast (DeepSeek V4 Flash)	DeepSeek	13.3B	45.3%	-27%
02	=	GLM 5.2	Zhipu	5B	17.1%	-33%
03	▼1	MiniMax M3	MiniMax	3.3B	11.1%	-29%
04	▲1	Kimi K3	Moonshot	2.9B	10%	+75%
05	=	SuperSmart (DeepSeek V4 Pro)	DeepSeek	2.8B	9.5%	-6%
06	·	Kimi K2.7 Coder	Moonshot	1.4B	4.6%	-72%
07	=	Qwen3.8 Max	Qwen	538.9M	1.8%	-28%
08	·	Qwen3.7 Max	Qwen	118.9M	0.4%	New
09	·	Qwen3.6 Flash	Qwen	39M	0.1%	New
10	·	Qwen3.7 Plus	Qwen	5.9M	0%	-94%
11	·	GLM 5.1	Zhipu	67.3K	0%	New

Session cost.

Average cost per session.

COST / SESSIONTOKENS / SESSION

$0.001855SuperFast (DeepSeek V4 Flash)

$0.004343Qwen3.6 Flash

$0.006274Qwen3.8 Max

$0.008843GLM 5.1

$0.0162MiniMax M3

$0.0238Kimi K2.7 Coder

$0.0255Qwen3.7 Plus

$0.0327SuperSmart (DeepSeek V4 Pro)

$0.0348GLM 5.2

$0.0376Kimi K3

$0.0463Qwen3.7 Max

Token cost.

Fixed pricing per 1M tokens (input / output).

COST / 1M TOKENS (IN → OUT)

$0.130 / $0.250SuperFast (DeepSeek V4 Flash)

$0.400 / $1.000SuperSmart (DeepSeek V4 Pro)

$0.600 / $2.400Minimax M3

$0.800 / $3.200Qwen3.7 Plus

$0.894 / $3.713Kimi K2.7 Code

$0.660 / $3.961Qwen3.6 Flash

$1.100 / $3.851GLM 5.2

$1.720 / $5.050Qwen3.8 Max

$3.000 / $9.000Kimi K3

Cache hits.

Share of prompt tokens served from cache.

96.9% of prompt tokens were
served from cache

96.2% median model hit rate

6 models reporting

Cached share of prompt tokens

97.8%SuperFast (DeepSeek V4 Flash)

97.7%SuperSmart (DeepSeek V4 Pro)

96.7%GLM 5.2

95.8%Qwen3.8 Max

94.3%Kimi K3

93.9%MiniMax M3

[*] dashed marker = fleet average · only requests with cached tokens are counted; models without cache hits are hidden.