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	gpt-6-astraOpenAI	83,296	4.2B	$92.372	−88%	72.3%	20.66s	19.1t/s
2	claude-opus-5-5Anthropic	38,661	2B	$0.00	−21%	68.7%	24.96s	21.0t/s
3	claude-sonnet-5Anthropic	12,525	365.6M	$0.12	−94%	66%	19.98s	17.6t/s
4	gpt-5.6-lunaOpenAI	10,940	287.8M	$0.4216	−69%	74.3%	16.54s	10.5t/s
5	gpt-5.6-solOpenAI	10,470	570.9M	$0.2685	−84%	73.4%	18.54s	18.6t/s
6	claude-haiku-4-5-20251001Anthropic	7,236	34.2M	$0.00	−98%	65.1%	26.77s	24.0t/s
7	claude-opus-5Anthropic	6,516	268.4M	$0.034	−92%	68.3%	26.47s	20.0t/s
8	claude-fable-5-1Anthropic	5,950	230.2M	$0.0016	−85%	70.3%	23.42s	19.9t/s
9	gpt-5.6-terraOpenAI	3,460	134.9M	$0.0006	−91%	73.2%	15.60s	18.8t/s
10	claude-opus-4-8Anthropic	2,250	85.2M	$0.0001	−96%	67.2%	21.55s	16.9t/s
11	gemini-3.8-flashGoogle	2,026	91.5M	$0.00	−95%	76.8%	20.47s	17.1t/s
12	Ddeepseek-v4-flashDeepSeek	1,976	57.3M	$0.0002	—	74.6%	19.11s	19.6t/s
13	codex-auto-reviewOpenAI	1,892	90.1M	$2.545	−93%	86.8%	16.02s	11.0t/s
14	claude-fable-5Anthropic	1,595	79.4M	$0.00	−81%	73.1%	18.91s	18.3t/s
15	gpt-5.4-miniOpenAI	1,350	3M	$0.00	−92%	73.2%	23.92s	27.4t/s
16	Ddeepseek-v4-proDeepSeek	704	26.4M	$0.0002	—	67%	21.08s	17.0t/s
17	Zglm-5.3Z.AI	617	27.5M	$0.0007	—	69%	15.01s	21.1t/s
18	gpt-5.4OpenAI	384	19M	$0.0001	−98%	70.1%	28.19s	12.0t/s
19	gpt-5.5OpenAI	330	1.5M	$0.001	−95%	33.9%	23.72s	17.8t/s
20	grok-4.7xAI	211	0	$0.00	−89%	0%	—	—
21	gemini-3.7-flashGoogle	200	1.1M	$0.0001	−95%	43%	18.22s	15.7t/s
22	gpt-oss-120bOpenAI	190	817.6K	$0.00	—	42.6%	24.01s	22.5t/s
23	gemini-3.1-flash-lite-imageGoogle	122	0	$0.00	−95%	0%	—	—