Rankings

Which model performs best

Speed, reliability and savings for each catalog model. Usage statistics update as API requests come in.

#	Model	Requests	Tokens	Spending	Savings	Uptime	Latency	TPS
1	claude-sonnet-5Anthropic	218,238	11.4B	$0.00	−95%	91.6%	20.65s	20.9t/s
2	gpt-5.6-lunaOpenAI	209,796	526M	$0.0003	−69%	75%	9.65s	13.7t/s
3	claude-opus-5-5Anthropic	190,692	15.4B	$0.00	−21%	91.4%	23.56s	22.5t/s
4	Ddeepseek-v4-flashDeepSeek	110,850	3.8B	$0.00	—	84.5%	18.68s	22.0t/s
5	claude-fable-5-1Anthropic	110,697	6.9B	$0.0002	−95%	83.1%	25.90s	24.0t/s
6	gpt-6.1-solOpenAI	60,845	4.6B	$0.00	−89%	88.1%	24.56s	27.2t/s
7	gpt-6-astraOpenAI	56,319	4.2B	$0.00	−88%	87.4%	31.41s	31.5t/s
8	Ddeepseek-v4-proDeepSeek	37,081	1.7B	$0.013	—	86.6%	21.92s	28.8t/s
9	claude-haiku-4-5-20251001Anthropic	36,432	711.5M	$0.00	−95%	85%	13.39s	13.6t/s
10	claude-opus-5Anthropic	35,690	2.4B	$0.0019	−95%	89%	21.82s	24.4t/s
11	gpt-5.4-miniOpenAI	31,694	236.3M	$0.00	−92%	66.9%	10.58s	18.6t/s
12	claude-fable-5Anthropic	24,441	1.5B	$0.0306	−95%	85.6%	23.94s	24.6t/s
13	gemini-3.8-flashGoogle	16,074	602.9M	$0.00	−95%	89.8%	31.64s	31.2t/s
14	grok-4.7xAI	13,607	120.9M	$0.00	−89%	55.9%	31.90s	48.4t/s
15	claude-opus-4-8Anthropic	12,610	633.6M	$0.00	−95%	85.3%	17.05s	16.1t/s
16	gpt-6-solOpenAI	11,426	492M	$0.00	−88%	85.6%	67.89s	44.7t/s
17	gpt-5.6-solOpenAI	10,562	730.4M	$0.00	−95%	87.1%	23.55s	27.5t/s
18	gpt-6-lunaOpenAI	9,359	375.7M	$0.0005	−69%	79.5%	18.69s	23.4t/s
19	gpt-5.6-terraOpenAI	7,198	186.3M	$0.00	−91%	85.1%	22.40s	33.1t/s
20	Zglm-5.3Z.AI	5,900	191.2M	$0.00	—	88.6%	27.41s	32.9t/s
21	gemini-3.7-flashGoogle	3,578	31.1M	$0.00	−95%	85.7%	32.02s	33.3t/s
22	codex-auto-reviewOpenAI	3,568	95.4M	$0.00	−95%	94%	13.12s	16.7t/s
23	gpt-5.4OpenAI	3,303	58.7M	$0.00	−95%	86.5%	12.24s	13.2t/s
24	gpt-5.5OpenAI	3,213	32.3M	$0.00	−95%	82.4%	13.34s	15.6t/s
25	gpt-oss-120bOpenAI	2,650	42.7M	$0.00	—	83.8%	9.55s	10.5t/s
26	gemini-3.1-flash-imageGoogle	180	0	$0.00	−95%	0%	—	—