Coding

React Native Evals leaderboard

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

As of 2026-10-07, the highest measured score on React Native Evals is 96.1% by Composer 2. 151 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#	Model	Score	Source
1	Claude Fable 5	100.0%	estimated ± 2.4 pp, low confidence
2	Claude Opus 5	100.0%	estimated ± 2.4 pp, low confidence
3	DeepSeek V4 Pro 0813	100.0%	estimated ± 2.4 pp, low confidence
4	GLM-5.3	100.0%	estimated ± 2.4 pp, low confidence
5	GLM-5.3-Flash	100.0%	estimated ± 2.4 pp, low confidence
6	GPT-5.6 Luna	100.0%	estimated ± 2.4 pp, low confidence
7	GPT-5.6 Sol	100.0%	estimated ± 2.4 pp, low confidence
8	GPT-5.6 Terra	100.0%	estimated ± 2.4 pp, low confidence
9	Grok 4.6	100.0%	estimated ± 2.4 pp, low confidence
10	Kimi K3	100.0%	estimated ± 2.4 pp, low confidence
11	DeepSeek V4 Flash 0731	97.4%	estimated ± 2.4 pp, low confidence
12	Claude Opus 4.8	97.0%	estimated ± 2.4 pp, low confidence
13	Composer 2	96.1%	measured
14	Composer 2 Fast	94.9%	measured
15	Grok 4.5	93.1%	estimated ± 2.4 pp, low confidence
16	Muse Spark 1.2	93.1%	estimated ± 2.4 pp, low confidence
17	Qwen3.8-27B	92.0%	estimated ± 2.4 pp, low confidence
18	Qwen3.8 Max	91.2%	estimated ± 2.4 pp, low confidence
19	Gemini 4 Argon	86.5%	estimated ± 3.0 pp, low confidence
20	GLM-5.2	86.1%	estimated ± 2.4 pp, low confidence
21	Mistral Large 4	85.4%	estimated ± 3.0 pp, low confidence
22	GPT-5.4	85.3%	measured
23	Inkling-Small	85.1%	estimated ± 2.4 pp, medium confidence
24	Muse Spark 1.1	84.7%	estimated ± 2.4 pp, medium confidence
25	GPT-5.5	84.7%	measured
26	Claude Fable 5.1	84.6%	estimated ± 3.5 pp, medium confidence
27	Claude Opus 4.6	84.1%	measured
28	Claude Opus 5.5	83.8%	estimated ± 4.7 pp, low confidence
29	GPT-6 Astra	83.4%	estimated ± 3.5 pp, medium confidence
30	Muse Spark 1.3	83.1%	estimated ± 3.5 pp, medium confidence
31	Claude Opus 4.7	82.8%	measured
32	Gemini 3.7 Flash	82.6%	estimated ± 2.4 pp, medium confidence
33	Claude Opus 4.7 (Adaptive)	82.6%	estimated ± 3.5 pp, high confidence
34	Qwen3.8-Flash-Next	82.4%	estimated ± 3.5 pp, high confidence
35	GPT-5.2	82.2%	estimated ± 3.0 pp, medium confidence
36	Claude Opus 4.6 (Adaptive)	82.2%	estimated ± 3.0 pp, medium confidence
37	Qwen3.8 Max Preview	82.1%	estimated ± 3.5 pp, high confidence
38	Claude Sonnet 5.5	81.3%	estimated ± 4.7 pp, low confidence
39	MiMo-V2.6-Pro	81.3%	estimated ± 4.7 pp, low confidence
40	Gemini 3.8 Flash	81.3%	estimated ± 2.4 pp, medium confidence
41	Claude Sonnet 5	80.6%	estimated ± 2.4 pp, medium confidence
42	Composer 2.5	80.6%	estimated ± 2.4 pp, medium confidence
43	Gemini 3.6 Flash	80.6%	estimated ± 2.4 pp, medium confidence
44	Claude Sonnet 4.6	80.6%	measured
45	Step 5 Preview	80.5%	estimated ± 4.7 pp, low confidence
46	GPT-6 Sol	79.9%	estimated ± 4.7 pp, medium confidence
47	Grok 4.7	79.8%	estimated ± 4.7 pp, medium confidence
48	Quasar 438B	79.5%	estimated ± 3.5 pp, high confidence
49	Apodex 1.1	79.3%	estimated ± 3.5 pp, high confidence
50	Apodex 1.1 Mini	79.3%	estimated ± 3.5 pp, high confidence
51	Gemini 3.5 Flash	79.3%	estimated ± 2.4 pp, medium confidence
52	Gemini 3.1 Pro	78.9%	measured
53	Hy3	78.8%	estimated ± 3.5 pp, high confidence
54	Hy3 Preview	78.8%	estimated ± 3.5 pp, high confidence
55	Claude Haiku 5.5	78.8%	estimated ± 4.7 pp, medium confidence
56	GPT-6 Luna	78.6%	estimated ± 4.7 pp, medium confidence
57	GPT-6.1 Sol	78.5%	estimated ± 4.7 pp, medium confidence
58	Ling 3.1 Flash	78.4%	estimated ± 4.7 pp, medium confidence
59	Kimi K2.7 Code	78.3%	estimated ± 2.4 pp, medium confidence
60	Qwen3.7 Plus	78.1%	estimated ± 3.5 pp, high confidence
61	GPT-5.3 Codex	78.0%	estimated ± 2.4 pp, medium confidence
62	DeepSeek V4.1 Flash	77.5%	estimated ± 4.7 pp, medium confidence
63	Inkling	77.3%	estimated ± 2.4 pp, medium confidence
64	MiMo-V2.6-Flash	77.2%	estimated ± 4.7 pp, medium confidence
65	Kimi K2.5	77.2%	measured
66	Ling 3.0 Flash FP8	76.8%	estimated ± 3.5 pp, high confidence
67	MiMo-V2-Flash	76.6%	estimated ± 3.5 pp, high confidence
68	Muse Glimmer 30B	76.4%	estimated ± 3.5 pp, high confidence
69	Qwen3.5-122B-A10B	75.6%	estimated ± 3.5 pp, high confidence
70	GLM-5.1	75.4%	estimated ± 2.4 pp, medium confidence
71	Gemma 4 31B	75.2%	measured
72	Kimi K2.6	75.0%	estimated ± 2.4 pp, medium confidence
73	GLM-5	74.8%	measured
74	Qwen3.6-35B-A3B	74.6%	estimated ± 3.5 pp, high confidence
75	Solar Pro 4	74.4%	estimated ± 4.7 pp, medium confidence
76	Ling 3.0 Flash VL	74.2%	estimated ± 4.7 pp, medium confidence
77	o1	74.1%	estimated ± 3.5 pp, high confidence
78	Step 3.7 Flash	74.0%	estimated ± 3.5 pp, high confidence
79	Gemma 4 26B A4B	74.0%	estimated ± 3.5 pp, high confidence
80	Nemotron 3 Super 100B	73.6%	estimated ± 3.5 pp, high confidence
81	K-EXAONE 2.0	73.3%	estimated ± 4.7 pp, medium confidence
82	Gemini 3.5 Flash-Lite	73.2%	estimated ± 2.4 pp, medium confidence
83	Gemini 3 Flash	73.2%	estimated ± 2.4 pp, medium confidence
84	MiniMax M3	73.2%	estimated ± 2.4 pp, medium confidence
85	A.X K2	72.8%	estimated ± 4.7 pp, medium confidence
86	o1-preview	72.6%	estimated ± 3.5 pp, high confidence
87	Grok 4	72.6%	measured
88	Muse Spark	72.2%	estimated ± 2.4 pp, medium confidence
89	K-Exaone	72.1%	estimated ± 3.5 pp, high confidence
90	GPT-5.1	72.1%	estimated ± 3.0 pp, low confidence
91	DeepSeek V3 0324	72.0%	estimated ± 4.7 pp, medium confidence
92	North Mini Code	71.9%	estimated ± 4.7 pp, medium confidence
93	Gemma 4 12B	71.9%	estimated ± 3.5 pp, high confidence
94	Mercury 2.5	71.8%	estimated ± 4.7 pp, medium confidence
95	MiMo-V2.5-Pro	71.6%	estimated ± 2.4 pp, medium confidence
96	GPT-OSS 120B	71.6%	measured
97	DeepSeek V3.2	71.5%	measured
98	Granite 4.2 30B	71.5%	estimated ± 4.7 pp, medium confidence
99	MiniMax M2.7	71.4%	measured
100	GLM-5 (Reasoning)	71.2%	estimated ± 3.0 pp, low confidence
101	Command A+	71.1%	estimated ± 3.5 pp, high confidence
102	GPT-OSS 20B	71.0%	measured
103	Nemotron 3.5 Lightning 30B A3B NVFP4	70.8%	estimated ± 3.5 pp, high confidence
104	Mistral Small 4	70.8%	estimated ± 3.5 pp, high confidence
105	Mistral Small 4 (Reasoning)	70.8%	estimated ± 3.5 pp, high confidence
106	Qwen3.6 Plus	70.7%	estimated ± 2.4 pp, low confidence
107	Claude Sonnet 4.5 Thinking	70.7%	estimated ± 3.0 pp, low confidence
108	Trinity-Large-Preview	70.5%	estimated ± 3.5 pp, high confidence
109	Trinity-Large-Thinking	70.5%	estimated ± 3.5 pp, high confidence
110	Ling 2.6 Flash	70.4%	estimated ± 3.5 pp, high confidence
111	GPT-5.1-Codex-Max	70.3%	estimated ± 3.0 pp, low confidence
112	GPT-5.4 mini	70.1%	estimated ± 2.4 pp, low confidence
113	Gemini 1.5 Pro	70.0%	estimated ± 3.5 pp, high confidence
114	DeepSeek V3	69.9%	estimated ± 3.5 pp, high confidence
115	Qwen 3.6 Max (preview)	69.8%	estimated ± 2.4 pp, low confidence
116	Granite 4.2 8B	69.7%	estimated ± 3.5 pp, high confidence
117	GPT-4 Turbo	69.5%	estimated ± 3.5 pp, high confidence
118	GPT-5.2-Codex	69.2%	estimated ± 2.4 pp, low confidence
119	GPT-4.1 mini	69.2%	estimated ± 3.5 pp, medium confidence
120	Mistral Large 3	69.1%	estimated ± 3.5 pp, medium confidence
121	Claude Opus 4.5 Thinking	69.0%	estimated ± 3.0 pp, low confidence
122	Claude 3 Opus	69.0%	estimated ± 3.5 pp, medium confidence
123	Grok 4.20	68.9%	estimated ± 2.4 pp, low confidence
124	GPT-5 (high)	68.6%	estimated ± 3.0 pp, low confidence
125	Llama 4 Maverick	68.2%	estimated ± 3.5 pp, medium confidence
126	Grok 4.3	67.7%	estimated ± 2.4 pp, low confidence
127	Celeris-1	67.7%	estimated ± 3.5 pp, medium confidence
128	Nemotron 3 Nano 30B	67.7%	estimated ± 3.5 pp, medium confidence
129	Nemotron 3 Nano Omni 30B A3B	67.5%	estimated ± 3.5 pp, medium confidence
130	MiMo-V2.5	67.1%	estimated ± 2.4 pp, low confidence
131	Ultravox v0.6 Llama 3.3 70B	67.1%	estimated ± 3.5 pp, medium confidence
132	GPT-4o mini	66.9%	estimated ± 3.5 pp, medium confidence
133	GPT-4.1 nano	66.9%	estimated ± 3.5 pp, medium confidence
134	MiniCPM5-2B	66.6%	estimated ± 4.7 pp, low confidence
135	Gemma 3 27B	66.6%	estimated ± 3.5 pp, medium confidence
136	Gemma 4 E4B	66.4%	estimated ± 3.5 pp, medium confidence
137	Solar Pro 3	66.3%	estimated ± 4.7 pp, low confidence
138	Granite 4.2 3B	66.2%	estimated ± 4.7 pp, low confidence
139	Llama 4 Scout	66.1%	estimated ± 3.5 pp, medium confidence
140	LFM2.5-2.6B	66.0%	estimated ± 3.5 pp, medium confidence
141	Kimi K2.5 (Reasoning)	66.0%	estimated ± 3.0 pp, low confidence
142	Gemma 4 E2B	65.9%	estimated ± 3.5 pp, medium confidence
143	Ling 3.0 Tiny	65.7%	estimated ± 4.7 pp, low confidence
144	Qwen3.6-27B	65.7%	estimated ± 2.4 pp, low confidence
145	GPT-5.4 nano	65.4%	estimated ± 2.4 pp, low confidence
146	GLM-4.7	64.8%	estimated ± 2.4 pp, low confidence
147	Nemotron 3 Ultra	64.2%	estimated ± 2.4 pp, low confidence
148	Qwen3.7 Max	63.9%	estimated ± 2.4 pp, low confidence
149	Qwen3.5 Plus	63.9%	estimated ± 3.0 pp, low confidence
150	MiniMax M2.5	62.7%	estimated ± 3.0 pp, low confidence
151	Gemini 3 Pro	61.9%	estimated ± 3.0 pp, low confidence
152	GPT-5 mini	61.7%	estimated ± 3.0 pp, low confidence
153	Claude Haiku 4.5	60.9%	estimated ± 2.4 pp, low confidence
154	Mistral Medium 3.5 128B	60.6%	estimated ± 2.4 pp, low confidence
155	GPT-5.1-Codex	60.1%	estimated ± 3.0 pp, low confidence
156	Ling 3.0 Flash	59.0%	estimated ± 2.4 pp, low confidence
157	Qwen3.5 Flash	57.9%	estimated ± 2.4 pp, low confidence
158	Claude Haiku 4.5 Thinking	57.0%	estimated ± 3.0 pp, low confidence
159	Gemini 3.1 Flash-Lite	55.8%	estimated ± 2.4 pp, low confidence
160	Laguna M.1	49.3%	estimated ± 2.4 pp, low confidence
161	Laguna XS.2	46.5%	estimated ± 2.4 pp, low confidence
162	Gemini 2.5 Pro	45.6%	estimated ± 2.4 pp, low confidence
163	DeepSeek V3.2 (Thinking)	38.5%	estimated ± 3.0 pp, low confidence
164	Qwen3 Max	30.4%	estimated ± 3.0 pp, low confidence
165	GLM-4.6	27.9%	estimated ± 3.0 pp, low confidence
166	Grok 4.1 Fast (Reasoning)	13.2%	estimated ± 3.0 pp, low confidence
167	Grok 4 Fast (Reasoning)	0.0%	estimated ± 3.0 pp, low confidence

React Native Evals leaderboard

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General