The leaders, at a glance

AI reasoning model ranking

Compare AI reasoning models and measured effort settings across published reasoning benchmarks.

See the full ranking ↓How configurations are ranked

Reasoning leaders

Experimental comparison score · 0–100

For the selected profile and sources. Scores are experimental; small gaps may not be meaningful. How scoring works · Effort evidence collected 2026-09-12

451unique modelsEach model counted once across effort levels 492model configurationsDocumented model + effort settings, including unranked

109aggregated sourcesOriginal reports and benchmark leaderboards

Explore all 109 sources Independent evaluations and lab reports

Distinct source pages with published results in the catalog. Multiple results and page sections count as one source.

Every ranked row is a measured configuration. The default view shows the highest-scoring supported configuration per model. In Value, the budget is applied first. Select All efforts for the full comparison. How effort ranking works.

Customize ranking

Full ranking

217 entries

Custom capability ranking of measured model configurations. Scores are experimental; missing results remain unknown.
Rank	Model	Score	Capability coverage	analyst-agent	briefcase	aime	apex-agents	aa-lcr	automationbench	critpt	enterprise-ops	gdp-pdf	gdpval	gmmlu	gpqa	harvey	hle	ifbench	itbench	livecodebench	math500	mmlu-pro	mmmu-pro	omniscience	scicode	tau-bench	terminal-bench	arc-agi	deepswe	swe-bench-pro	arc-agi-3	frontiercode	enigma-eval	swe-atlas-qna	swe-atlas-refactoring	swe-atlas-test-writing	livebench-math	livebench-coding	livebench-language	livebench-data	multi-swe-bench	livebench-reasoning	livebench-instructions	swe-rebench	epoch-game-puzzles	frontiermath	simpleqa	vals-finance-agent	vals-code-migration	vals-legal-research	vals-excel-modeling	vibe-code	chartography	osworld	charxiv	swe-bench-multilingual	deepsearchqa	officeqa	mcp-atlas	toolathlon	mrcr	browsecomp
126	GPT-5 Nano · HighOpenAI Measured effort: High	6.8	Eligible · 100%15 published · 5 families · 1 selected capabilities	—	—	83.8108/108 peersIndependent evidence	—	46.5197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	50.163/63 peersIndependent evidence	41.5212/212 peersIndependent evidence	—	50.0208/208 peersIndependent evidence	81.1178/178 peersIndependent evidence	—	90.1112/112 peersIndependent evidence	—	56.9114/114 peersIndependent evidence	37.6114/114 peersIndependent evidence	51.2198/198 peersIndependent evidence	—	47.5171/171 peersIndependent evidence	50.6169/169 peersIndependent evidence	26.038/38 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
127	Mistral Medium 3.5 · ReasoningMistral Measured effort: Reasoning	6.7	Eligible · 100%22 published · 3 families · 1 selected capabilities	41.528/28 peersIndependent evidence	39.487/87 peersIndependent evidence	—	—	46.9198/198 peersIndependent evidence	47.989/89 peersIndependent evidence	25.2200/200 peersIndependent evidence	43.641/41 peersIndependent evidence	33.388/88 peersIndependent evidence	53.6123/123 peersIndependent evidence	—	55.1212/212 peersIndependent evidence	22.537/37 peersIndependent evidence	62.5208/208 peersIndependent evidence	81.2178/178 peersIndependent evidence	—	—	—	—	51.8114/114 peersIndependent evidence	67.4198/198 peersIndependent evidence	37.791/91 peersIndependent evidence	75.6202/202 peersIndependent evidence	55.6201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
128	GLM-4.6V · ReasoningZ.ai Measured effort: Reasoning	6.6	Eligible · 100%14 published · 4 families · 1 selected capabilities	—	—	87.9108/108 peersIndependent evidence	—	50.9197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	42.763/63 peersIndependent evidence	49.4212/212 peersIndependent evidence	—	50.9208/208 peersIndependent evidence	15.4178/178 peersIndependent evidence	—	10.5112/112 peersIndependent evidence	—	64.3114/114 peersIndependent evidence	17.6114/114 peersIndependent evidence	33.7198/198 peersIndependent evidence	—	39.4171/171 peersIndependent evidence	54.0169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
129	Magistral Small 1.2 · ReasoningMistral Measured effort: Reasoning	6.5	Eligible · 100%18 published · 4 families · 1 selected capabilities	—	—	77.8108/108 peersIndependent evidence	—	22.4197/197 peersIndependent evidence	2.889/89 peersIndependent evidence	55.9200/200 peersIndependent evidence	—	—	15.1123/123 peersIndependent evidence	28.363/63 peersIndependent evidence	36.2212/212 peersIndependent evidence	—	39.2208/208 peersIndependent evidence	53.2178/178 peersIndependent evidence	—	78.0112/112 peersIndependent evidence	—	52.5114/114 peersIndependent evidence	29.7114/114 peersIndependent evidence	19.5198/198 peersIndependent evidence	—	22.3202/202 peersIndependent evidence	21.2201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
130	Qwen3-32B · ReasoningQwen Measured effort: Reasoning	6.5	Eligible · 100%21 published · 5 families · 1 selected capabilities	—	9.287/87 peersIndependent evidence	71.3108/108 peersIndependent evidence	—	5.8197/197 peersIndependent evidence	22.489/89 peersIndependent evidence	55.7200/200 peersIndependent evidence	—	25.388/88 peersIndependent evidence	17.0123/123 peersIndependent evidence	—	39.0212/212 peersIndependent evidence	—	43.4208/208 peersIndependent evidence	31.4178/178 peersIndependent evidence	—	58.3112/112 peersIndependent evidence	74.460/60 peersIndependent evidence	63.8114/114 peersIndependent evidence	—	42.3198/198 peersIndependent evidence	23.891/91 peersIndependent evidence	27.6202/202 peersIndependent evidence	19.9201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
131	Qwen3-VL-30B-A3B-Thinking · ReasoningQwen Measured effort: Reasoning	6.2	Eligible · 100%12 published · 4 families · 1 selected capabilities	—	—	81.1108/108 peersIndependent evidence	—	—	—	25.4200/200 peersIndependent evidence	—	—	—	—	49.1212/212 peersIndependent evidence	—	48.7208/208 peersIndependent evidence	54.7178/178 peersIndependent evidence	—	73.7112/112 peersIndependent evidence	—	68.5114/114 peersIndependent evidence	39.8114/114 peersIndependent evidence	40.1198/198 peersIndependent evidence	—	19.8171/171 peersIndependent evidence	30.7169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
132	Qwen3.5-4B · ReasoningQwen Measured effort: Reasoning	6.1	Eligible · 100%12 published · 3 families · 1 selected capabilities	—	—	—	—	64.1197/197 peersIndependent evidence	—	25.0200/200 peersIndependent evidence	—	—	—	—	62.4212/212 peersIndependent evidence	—	53.0208/208 peersIndependent evidence	64.6178/178 peersIndependent evidence	—	—	—	—	53.1114/114 peersIndependent evidence	26.9198/198 peersIndependent evidence	—	63.5202/202 peersIndependent evidence	52.7201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
133	GLM-4.5-Air · ReasoningZ.ai Measured effort: Reasoning	6.0	Eligible · 100%13 published · 5 families · 1 selected capabilities	—	—	78.7108/108 peersIndependent evidence	—	48.5197/197 peersIndependent evidence	—	25.5200/200 peersIndependent evidence	—	—	—	—	51.4212/212 peersIndependent evidence	—	42.9208/208 peersIndependent evidence	34.5178/178 peersIndependent evidence	—	70.2112/112 peersIndependent evidence	76.260/60 peersIndependent evidence	72.5114/114 peersIndependent evidence	—	34.5198/198 peersIndependent evidence	—	56.1171/171 peersIndependent evidence	63.7169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
134	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 · ReasoningNVIDIA Measured effort: Reasoning	5.8	Eligible · 100%16 published · 3 families · 1 selected capabilities	—	36.087/87 peersIndependent evidence	—	—	41.0198/198 peersIndependent evidence	17.689/89 peersIndependent evidence	25.5200/200 peersIndependent evidence	0.041/41 peersIndependent evidence	33.688/88 peersIndependent evidence	47.3123/123 peersIndependent evidence	—	54.3212/212 peersIndependent evidence	—	54.8208/208 peersIndependent evidence	—	—	—	—	—	—	23.7198/198 peersIndependent evidence	15.991/91 peersIndependent evidence	28.9112/112 peersIndependent evidence	41.7125/125 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
135	MiniMax-M1-80k · ReasoningMiniMax Measured effort: Reasoning	5.7	Eligible · 100%13 published · 5 families · 1 selected capabilities	—	—	58.5108/108 peersIndependent evidence	—	58.0197/197 peersIndependent evidence	—	25.5200/200 peersIndependent evidence	—	—	—	—	45.3212/212 peersIndependent evidence	—	47.8208/208 peersIndependent evidence	46.9178/178 peersIndependent evidence	—	76.4112/112 peersIndependent evidence	91.260/60 peersIndependent evidence	75.2114/114 peersIndependent evidence	—	62.3198/198 peersIndependent evidence	—	44.0171/171 peersIndependent evidence	20.9169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
136	Qwen3-235B-A22B · ReasoningQwen Measured effort: Reasoning	5.6	Eligible · 100%13 published · 5 families · 1 selected capabilities	—	—	79.8108/108 peersIndependent evidence	—	5.8197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	—	46.6212/212 peersIndependent evidence	—	57.0208/208 peersIndependent evidence	36.8178/178 peersIndependent evidence	—	64.9112/112 peersIndependent evidence	58.360/60 peersIndependent evidence	82.4114/114 peersIndependent evidence	—	47.6198/198 peersIndependent evidence	—	28.1171/171 peersIndependent evidence	33.2169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
137	Qwen3-Coder-Next · Non-reasoningQwen Measured effort: Non-reasoning	5.4	Eligible · 100%18 published · 3 families · 1 selected capabilities	—	27.887/87 peersIndependent evidence	—	—	48.8197/197 peersIndependent evidence	24.389/89 peersIndependent evidence	25.4200/200 peersIndependent evidence	—	29.588/88 peersIndependent evidence	40.9123/123 peersIndependent evidence	38.363/63 peersIndependent evidence	51.8212/212 peersIndependent evidence	—	51.5208/208 peersIndependent evidence	29.7178/178 peersIndependent evidence	—	—	—	—	—	32.4198/198 peersIndependent evidence	24.991/91 peersIndependent evidence	45.5202/202 peersIndependent evidence	44.0201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
138	Solar Pro 3 · ReasoningUpstage Measured effort: Reasoning	5.3	Eligible · 100%18 published · 3 families · 1 selected capabilities	—	14.887/87 peersIndependent evidence	—	—	24.2198/198 peersIndependent evidence	10.789/89 peersIndependent evidence	25.5200/200 peersIndependent evidence	—	23.888/88 peersIndependent evidence	24.4123/123 peersIndependent evidence	—	49.4212/212 peersIndependent evidence	—	53.0208/208 peersIndependent evidence	85.2178/178 peersIndependent evidence	—	—	—	—	—	46.1198/198 peersIndependent evidence	6.891/91 peersIndependent evidence	58.2202/202 peersIndependent evidence	30.7201/201 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
139	Qwen3-Next-80B-A3B-Instruct · Non-reasoningQwen Measured effort: Non-reasoning	5.2	Eligible · 100%12 published · 4 families · 1 selected capabilities	—	—	63.7108/108 peersIndependent evidence	—	52.5197/197 peersIndependent evidence	—	25.4200/200 peersIndependent evidence	—	—	—	—	52.7212/212 peersIndependent evidence	—	45.0208/208 peersIndependent evidence	41.7178/178 peersIndependent evidence	—	70.4112/112 peersIndependent evidence	—	77.9114/114 peersIndependent evidence	—	41.6198/198 peersIndependent evidence	—	22.8171/171 peersIndependent evidence	42.9169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
140	GLM-4.7-Flash · ReasoningZ.ai Measured effort: Reasoning	5.2	Eligible · 100%9 published · 3 families · 1 selected capabilities	—	—	—	—	43.0197/197 peersIndependent evidence	—	55.3200/200 peersIndependent evidence	—	—	—	—	27.8212/212 peersIndependent evidence	—	44.7208/208 peersIndependent evidence	73.1178/178 peersIndependent evidence	—	—	—	—	—	32.6198/198 peersIndependent evidence	—	100.0171/171 peersIndependent evidence	62.5169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
141	Step 3 VL 10B · ReasoningStepFun Measured effort: Reasoning	5.1	Eligible · 100%10 published · 3 families · 1 selected capabilities	—	—	—	—	5.7197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	—	44.4212/212 peersIndependent evidence	—	56.0208/208 peersIndependent evidence	61.3178/178 peersIndependent evidence	—	—	—	—	46.9114/114 peersIndependent evidence	19.0198/198 peersIndependent evidence	—	12.3171/171 peersIndependent evidence	30.4169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
142	Qwen3-4B-Thinking-2507 · ReasoningQwen Measured effort: Reasoning	5.1	Eligible · 100%12 published · 4 families · 1 selected capabilities	—	—	82.6108/108 peersIndependent evidence	—	39.9197/197 peersIndependent evidence	—	25.4200/200 peersIndependent evidence	—	—	—	—	38.5212/212 peersIndependent evidence	—	37.5208/208 peersIndependent evidence	60.6178/178 peersIndependent evidence	—	65.8112/112 peersIndependent evidence	—	42.9114/114 peersIndependent evidence	—	16.5198/198 peersIndependent evidence	—	30.6171/171 peersIndependent evidence	12.4169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
143	Qwen3-30B-A3B-Instruct-2507 · Non-reasoningQwen Measured effort: Non-reasoning	5.0	Eligible · 100%14 published · 5 families · 1 selected capabilities	—	—	63.9108/108 peersIndependent evidence	—	29.5197/197 peersIndependent evidence	—	25.5200/200 peersIndependent evidence	—	—	—	18.963/63 peersIndependent evidence	35.7212/212 peersIndependent evidence	—	41.4208/208 peersIndependent evidence	24.3178/178 peersIndependent evidence	—	55.8112/112 peersIndependent evidence	84.460/60 peersIndependent evidence	56.4114/114 peersIndependent evidence	—	26.4198/198 peersIndependent evidence	—	7.9171/171 peersIndependent evidence	34.2169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
144	Qwen3.5-omni-flash · Non-reasoningQwen Measured effort: Non-reasoning	4.9	Eligible · 100%10 published · 3 families · 1 selected capabilities	—	—	—	—	51.9197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	—	54.1212/212 peersIndependent evidence	—	44.1208/208 peersIndependent evidence	35.2178/178 peersIndependent evidence	—	—	—	—	50.6114/114 peersIndependent evidence	24.8198/198 peersIndependent evidence	—	76.6171/171 peersIndependent evidence	45.3169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
145	Qwen3-VL-235B-A22B-Instruct · Non-reasoningQwen Measured effort: Non-reasoning	4.9	Eligible · 100%13 published · 4 families · 1 selected capabilities	—	—	67.7108/108 peersIndependent evidence	—	35.5197/197 peersIndependent evidence	—	25.5200/200 peersIndependent evidence	—	—	—	—	47.4212/212 peersIndependent evidence	—	40.0208/208 peersIndependent evidence	48.6178/178 peersIndependent evidence	—	63.0112/112 peersIndependent evidence	—	80.2114/114 peersIndependent evidence	58.7114/114 peersIndependent evidence	54.3198/198 peersIndependent evidence	—	46.0171/171 peersIndependent evidence	39.1169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
146	Qwen3-VL-30B-A3B-Instruct · Non-reasoningQwen Measured effort: Non-reasoning	4.8	Eligible · 100%12 published · 4 families · 1 selected capabilities	—	—	70.0108/108 peersIndependent evidence	—	—	—	25.4200/200 peersIndependent evidence	—	—	—	—	44.7212/212 peersIndependent evidence	—	38.2208/208 peersIndependent evidence	24.8178/178 peersIndependent evidence	—	49.8112/112 peersIndependent evidence	—	50.8114/114 peersIndependent evidence	42.5114/114 peersIndependent evidence	29.8198/198 peersIndependent evidence	—	17.7171/171 peersIndependent evidence	34.3169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
147	GLM-4.5V · ReasoningZ.ai Measured effort: Reasoning	4.7	Eligible · 100%13 published · 4 families · 1 selected capabilities	—	—	71.3108/108 peersIndependent evidence	—	5.8197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	—	43.5212/212 peersIndependent evidence	—	38.5208/208 peersIndependent evidence	28.3178/178 peersIndependent evidence	—	63.8112/112 peersIndependent evidence	—	59.7114/114 peersIndependent evidence	19.4114/114 peersIndependent evidence	57.9198/198 peersIndependent evidence	—	24.9171/171 peersIndependent evidence	30.1169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
148	Qwen3-VL-32B-Instruct · Non-reasoningQwen Measured effort: Non-reasoning	4.5	Eligible · 100%12 published · 4 families · 1 selected capabilities	—	—	66.2108/108 peersIndependent evidence	—	—	—	25.4200/200 peersIndependent evidence	—	—	—	—	40.2212/212 peersIndependent evidence	—	40.8208/208 peersIndependent evidence	39.2178/178 peersIndependent evidence	—	54.5112/112 peersIndependent evidence	—	60.9114/114 peersIndependent evidence	48.4114/114 peersIndependent evidence	25.5198/198 peersIndependent evidence	—	37.0171/171 peersIndependent evidence	46.0169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
149	Solar Open 100B · ReasoningUpstage Measured effort: Reasoning	4.4	Eligible · 100%10 published · 3 families · 1 selected capabilities	—	—	—	—	42.4197/197 peersIndependent evidence	—	25.2200/200 peersIndependent evidence	—	—	—	50.663/63 peersIndependent evidence	35.4212/212 peersIndependent evidence	—	54.6208/208 peersIndependent evidence	68.3178/178 peersIndependent evidence	—	—	—	—	—	47.2198/198 peersIndependent evidence	—	57.6171/171 peersIndependent evidence	17.4169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
150	Qwen3-30B-A3B · ReasoningQwen Measured effort: Reasoning	4.3	Eligible · 100%13 published · 5 families · 1 selected capabilities	—	—	70.0108/108 peersIndependent evidence	—	5.9197/197 peersIndependent evidence	—	25.3200/200 peersIndependent evidence	—	—	—	—	32.1212/212 peersIndependent evidence	—	37.6208/208 peersIndependent evidence	45.1178/178 peersIndependent evidence	—	51.6112/112 peersIndependent evidence	74.360/60 peersIndependent evidence	55.7114/114 peersIndependent evidence	—	33.5198/198 peersIndependent evidence	—	31.7171/171 peersIndependent evidence	17.4169/169 peersIndependent evidence	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—

About this ranking

This profile gives hard reasoning all the weight. It compares identified configurations using their own measured results, rather than borrowing results from another effort setting. Rankings require support in this capability. Matching effort labels across providers do not imply equal compute budgets.

View the Capability ranking →

Behind the ranking

273 entries with published results · 217 ranked · 0 provisional estimates.

Explore seven capabilities, the original results, and gaps in the evidence.

Coding ranking Reasoning ranking Agentic ranking Compare capabilities ↗Published benchmarks ↗Evidence coverage ↗

Cite this ranking

Include the model and effort setting, the selected profile, and the dated data downloads. A shared link uses the latest data; it does not freeze a ranking in time.

UnifyBench ranking
Data updated: 12 Sept 2026
Mode: custom; evidence: documented; configurations: best
Weights: v3;agentic:0,hard_reasoning:100,coding:0,human_pref:0,knowledge:0,multimodal:0,long_context:0
https://unifybench.ai/rankings/reasoning?page=6

Catalog data · Effort measurements · Configuration scoring rules · Methodology

Catalog dated 12 Sept 2026. Effort evidence collected 12 Sept 2026. Collection dates are not evaluation dates.

What does the score mean?

The adjusted comparison score estimates performance against the same reference panel from matched published comparisons. Capability pools matched comparisons across families and checks aggregate breadth and independent corroboration. Custom profiles require support in every selected capability. Missing results never count as losses.

Family details show observed win shares against available reference peers. Those descriptive values are supporting evidence, not adjusted capability scores. Provider reports and independent results remain labeled. Matching reported settings do not establish equal inference effort, and score differences do not establish statistical significance.

Read the full methodology →