IOI

Vals AI's overall score for version 2 of the International Olympiad in Informatics benchmark.

37 models · 38 results · updated 2026-10-24 · vals.ai ↗

More from this benchmark

1 published measures

Time per task46 results · lower is better

Highlight models

Click a point on the chart to highlight that exact effort; click it again to remove it.

24 more in settings

Chart settings

37 of 37 selected

38 rows

#	Model	Effort	IOI	Subscription $ / task	API $ / task	Time
1	Gemini 4 Argon	high	100.0%	$1.89	$5.31	24 min
2	GPT-6 Astra	max	100.0%	$0.56	$6.50	32 min
3	GPT-6.1 Sol	max	96.9%	$0.15	$1.26	32 min
4	Opus 5.5	max	95.1%	$0.080	$5.26	17 min
5	GPT-5.6 Sol	max	91.2%	$0.44	$7.98	1.0 h
6	Fable 5.1	max	90.8%	$1.75	$11	30 min
7	GPT-5.6 Terra	max	87.6%	$0.48	$8.66	1.4 h
8	Opus 5	max	84.3%	$0.29	$16	57 min
9	Sonnet 5.5	max	83.1%	$0.11	$7.52	46 min
10	GPT-6 Sol	max	82.6%	$0.17	$2.70	31 min
11	Qwen 3.8 Max	max	68.9%	$9.47	$9.47	1.8 h
12	GLM 5.3	max	68.4%	$0.80	$7.67	1.8 h
13	Gemini 3.7 Flash	high	67.8%	$0.44	$3.40	21 min
14	GPT-5.6 Luna	max	61.8%	$0.054	$0.58	1.2 h
15	Grok 4.7	xhigh	57.7%	$1.18	$13	47 min
16	Gemini 3.8 Flash	high	56.9%	$0.16	$3.98	14 min
17	Muse Spark 1.3	max	56.6%	$0.86	$1.73	29 min
18	GPT-6 Luna	max	55.6%	$0.029	$0.20	36 min
19	GPT-5.3 Codex	xhigh	53.8%	$2.68	$2.68	30 min
20	GLM 5.3 Flash	max	52.5%	free	$0.43	1.5 h
21	DeepSeek V4 Pro 0813	max	51.6%	$2.15	$2.15	1.1 h
22	Kimi K3	max	48.9%	$1.09	$14	4.3 h
23	MiMo V2.6 Flash	high	47.7%	free	$0.22	1.8 h
24	Grok 4.6	high	47.6%	$0.84	$8.18	4.0 h
25	Haiku 5.5	max	47.3%	$0.018	$0.83	13 min
26	Mistral Large 4	high	45.3%	$5.39	$5.39	1.4 h
27	Sonnet 5	max	45.0%	$0.22	$13	60 min
28	Muse Spark 1.3	xhigh	43.9%	$1.18	$2.36	30 min
29	Grok 4.5	high	40.6%	$0.88	$5.64	1.4 h
30	DeepSeek V4.1 Flash	max	40.3%	$0.015	$0.24	18 min