benchgap
Coding

React Native Evals leaderboard

As of 2026-11-21, the highest measured score on React Native Evals is 96.1% by Composer 2. 216 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#ModelScoreSource
1Composer 296.1%measured
2Composer 2 Fast94.9%measured
3Claude Opus 586.9%estimated ± 2.7 pp, low confidence
4GLM-5.3-Flash86.7%estimated ± 4.7 pp, low confidence
5Claude Mythos 586.6%estimated ± 2.7 pp, low confidence
6Gemini 4 Argon86.5%estimated ± 3.0 pp, low confidence
7Claude Fable 586.4%estimated ± 2.7 pp, low confidence
8Mistral Large 485.4%estimated ± 3.0 pp, low confidence
9GPT-5.485.3%measured
10Ember-185.1%estimated ± 2.7 pp, low confidence
11GPT-5.584.7%measured
12Claude Fable 5.184.6%estimated ± 3.2 pp, medium confidence
13Claude Fable 5.284.2%estimated ± 4.7 pp, low confidence
14Claude Opus 4.684.1%measured
15Qwen3.8 Max83.9%estimated ± 4.7 pp, low confidence
16Claude Opus 5.583.8%estimated ± 4.7 pp, low confidence
17GPT-5.6 Sol83.6%estimated ± 3.2 pp, medium confidence
18Claude Opus 4.883.5%estimated ± 2.7 pp, low confidence
19GPT-6 Astra83.5%estimated ± 3.2 pp, medium confidence
20Grok 4.683.4%estimated ± 3.2 pp, medium confidence
21GPT-5.6 Terra83.4%estimated ± 3.2 pp, medium confidence
22Gemini 3.8 Flash83.3%estimated ± 3.2 pp, medium confidence
23Kimi K383.3%estimated ± 3.2 pp, medium confidence
24Gemini 3.7 Flash83.3%estimated ± 3.2 pp, medium confidence
25Muse Spark 1.383.2%estimated ± 3.2 pp, medium confidence
26GLM-5.382.9%estimated ± 3.2 pp, high confidence
27Claude Opus 4.782.8%measured
28Qwen3.8-Flash-Next82.5%estimated ± 3.2 pp, high confidence
29Ornith-1.5-397B82.3%estimated ± 2.7 pp, medium confidence
30Grok 4.582.3%estimated ± 3.2 pp, high confidence
31Muse Spark 1.282.3%estimated ± 3.2 pp, high confidence
32Claude Opus 4.6 (Adaptive)82.2%estimated ± 3.0 pp, medium confidence
33Qwen3.8 Max Preview82.2%estimated ± 3.2 pp, high confidence
34GPT-5.6 Luna82.1%estimated ± 3.2 pp, high confidence
35Muse Spark 1.182.1%estimated ± 3.2 pp, high confidence
36Muse Wonder Full82.0%estimated ± 4.7 pp, low confidence
37Claude Sonnet 582.0%estimated ± 2.7 pp, medium confidence
38GPT-5.3 Codex81.9%estimated ± 2.7 pp, medium confidence
39Gemini 3.6 Flash81.5%estimated ± 3.2 pp, high confidence
40GLM-5.281.4%estimated ± 3.2 pp, high confidence
41Claude Sonnet 5.581.3%estimated ± 4.7 pp, low confidence
42MiMo-V2.6-Pro81.3%estimated ± 4.7 pp, low confidence
43Composer 2.581.3%estimated ± 4.7 pp, medium confidence
44Kimi K3.1 Pro81.3%estimated ± 4.7 pp, low confidence
45Qwen3.8-27B81.2%estimated ± 3.2 pp, high confidence
46Gemini 3.5 Flash81.2%estimated ± 3.0 pp, medium confidence
47GPT-5.4 mini81.1%estimated ± 3.0 pp, medium confidence
48Muse Wonder81.0%estimated ± 4.7 pp, low confidence
49Ornith-1.0-397B80.7%estimated ± 2.7 pp, medium confidence
50Inkling 2 Small80.6%estimated ± 2.7 pp, medium confidence
51Claude Sonnet 4.680.6%measured
52Step 5 Preview80.5%estimated ± 4.7 pp, low confidence
53Beam 1.0 501B-A23B80.4%estimated ± 2.7 pp, medium confidence
54Muse Spark Open80.2%estimated ± 4.7 pp, medium confidence
55Claude Opus 4.580.1%estimated ± 2.7 pp, medium confidence
56Beam 1.0 501B-A23B NVFP480.0%estimated ± 2.7 pp, medium confidence
57DeepSeek V4 Pro79.9%estimated ± 2.7 pp, medium confidence
58GPT-6 Sol79.9%estimated ± 4.7 pp, medium confidence
59MiniMax M379.9%estimated ± 2.7 pp, medium confidence
60Qwen3.7 Max79.8%estimated ± 2.7 pp, medium confidence
61Grok 4.779.8%estimated ± 4.7 pp, medium confidence
62Inkling-Small79.7%estimated ± 2.7 pp, medium confidence
63Kimi K2.679.7%estimated ± 2.7 pp, medium confidence
64Beam 1.0 501B-A23B FP879.7%estimated ± 2.7 pp, medium confidence
65GPT-5.279.6%estimated ± 2.7 pp, medium confidence
66Quasar 438B79.5%estimated ± 3.2 pp, high confidence
67Apodex 1.1 Mini79.4%estimated ± 3.2 pp, high confidence
68Kimi K2.7 Code79.4%estimated ± 3.2 pp, high confidence
69Inkling 2 Mini79.3%estimated ± 2.7 pp, medium confidence
70MiMo-V2.5-Pro79.2%estimated ± 3.2 pp, high confidence
71DeepSeek V4 Flash 073179.2%estimated ± 2.7 pp, medium confidence
72Ornith-1.5-35B-A3B79.2%estimated ± 2.7 pp, medium confidence
73Qwen3.6 Plus79.1%estimated ± 2.7 pp, medium confidence
74BTL-478.9%estimated ± 2.7 pp, medium confidence
75dots3-note Preview78.9%estimated ± 2.7 pp, medium confidence
76Gemini 3.1 Pro78.9%measured
77Hy378.9%estimated ± 3.2 pp, high confidence
78GPT-6.1 Terra78.8%estimated ± 4.7 pp, medium confidence
79Claude Haiku 5.578.8%estimated ± 4.7 pp, medium confidence
80MiMo-V2-Pro78.7%estimated ± 2.7 pp, medium confidence
81GPT-6 Luna78.6%estimated ± 4.7 pp, medium confidence
82Apodex 1.178.6%estimated ± 2.7 pp, medium confidence
83Qwen3.7 Plus78.6%estimated ± 2.7 pp, medium confidence
84Inkling78.6%estimated ± 2.7 pp, medium confidence
85Mistral Medium 3.5 128B78.6%estimated ± 2.7 pp, medium confidence
86Muse Spark78.5%estimated ± 2.7 pp, medium confidence
87GPT-6.1 Sol78.5%estimated ± 4.7 pp, medium confidence
88Ling 3.1 Flash78.4%estimated ± 4.7 pp, medium confidence
89Claude Sonnet 4.578.4%estimated ± 2.7 pp, medium confidence
90Qwen3.6-27B78.4%estimated ± 2.7 pp, medium confidence
91GPT-5.2-Codex78.3%estimated ± 3.0 pp, medium confidence
92Qwen 3.6 Max (preview)78.2%estimated ± 4.7 pp, medium confidence
93Grok 4.2078.2%estimated ± 2.7 pp, medium confidence
94MiniMax M3.1 Flash78.1%estimated ± 4.7 pp, medium confidence
95Ripple-65B Preview78.1%estimated ± 2.7 pp, medium confidence
96Qwen3.5 397B77.9%estimated ± 2.7 pp, medium confidence
97Muse Glimmer 30B77.8%estimated ± 2.7 pp, medium confidence
98Ornith-1.0-35B77.7%estimated ± 2.7 pp, medium confidence
99DeepSeek V4.1 Flash 101777.5%estimated ± 4.7 pp, medium confidence
100MiMo-V2.577.3%estimated ± 4.7 pp, medium confidence
101MiMo-V2-Omni77.3%estimated ± 2.7 pp, medium confidence
102MiMo-V2.6-Flash77.2%estimated ± 4.7 pp, medium confidence
103Laguna M.177.2%estimated ± 2.7 pp, medium confidence
104Kimi K2.577.2%measured
105Claude 4.1 Opus77.2%estimated ± 2.7 pp, medium confidence
106Hy3 Preview77.1%estimated ± 2.7 pp, medium confidence
107Step 3.5 Flash77.1%estimated ± 2.7 pp, medium confidence
108Flint77.0%estimated ± 2.7 pp, medium confidence
109GLM-4.776.9%estimated ± 2.7 pp, medium confidence
110Ling 3.0 Flash76.8%estimated ± 3.2 pp, high confidence
111MAI-Thinking-176.7%estimated ± 2.7 pp, medium confidence
112MiMo-V2-Flash76.7%estimated ± 2.7 pp, medium confidence
113Qwen3.6-35B-A3B76.7%estimated ± 2.7 pp, medium confidence
114Claude Haiku 4.576.6%estimated ± 2.7 pp, medium confidence
115Gemini 3.5 Flash-Lite76.5%estimated ± 3.2 pp, high confidence
116Jais-3-MoVA-36B-A4B76.4%estimated ± 2.7 pp, medium confidence
117Claude 4 Sonnet76.4%estimated ± 2.7 pp, medium confidence
118MAI-Code-1.1-Flash76.3%estimated ± 2.7 pp, medium confidence
119Claude 4 Opus76.3%estimated ± 2.7 pp, medium confidence
120Jais-3-32B76.2%estimated ± 2.7 pp, medium confidence
121Qwen3.5-27B76.2%estimated ± 2.7 pp, medium confidence
122Qwen3.5-122B-A10B76.0%estimated ± 2.7 pp, medium confidence
123Nemotron 3 Ultra76.0%estimated ± 2.7 pp, medium confidence
124GLM-5.175.8%estimated ± 3.0 pp, medium confidence
125Nemotron 3.5 Super VL 120B A12B75.6%estimated ± 2.7 pp, medium confidence
126Nemotron 3.5 Super VL 120B A12B NVFP475.6%estimated ± 2.7 pp, medium confidence
127Laguna XS 2.175.5%estimated ± 2.7 pp, medium confidence
128Grok Code Fast 175.5%estimated ± 2.7 pp, medium confidence
129Ornith-1.5-9B75.4%estimated ± 2.7 pp, medium confidence
130Solar Pro 475.4%estimated ± 2.7 pp, medium confidence
131Solar Open 275.3%estimated ± 2.7 pp, medium confidence
132Jais-3-375B-A23B75.2%estimated ± 2.7 pp, medium confidence
133Jais-3-7B75.2%estimated ± 2.7 pp, medium confidence
134Gemma 4 31B75.2%measured
135Laguna XS.275.1%estimated ± 2.7 pp, medium confidence
136Qwen3-Coder 480B-A35B75.0%estimated ± 2.7 pp, medium confidence
137Ornith-1.0-9B74.9%estimated ± 2.7 pp, medium confidence
138GLM-574.8%measured
139Qwen3.5-35B-A3B74.8%estimated ± 2.7 pp, medium confidence
140Grok 4.374.7%estimated ± 3.2 pp, high confidence
141K-EXAONE 2.074.3%estimated ± 2.7 pp, medium confidence
142LongCat-Flash-Lite-Sparse74.3%estimated ± 2.7 pp, medium confidence
143Ling 3.0 Flash VL74.2%estimated ± 4.7 pp, medium confidence
144Qwen3.5 Flash74.1%estimated ± 4.7 pp, medium confidence
145o174.1%estimated ± 3.2 pp, high confidence
146Step 3.7 Flash74.0%estimated ± 3.2 pp, high confidence
147Gemma 4 26B A4B74.0%estimated ± 3.2 pp, high confidence
148DeepSeek V3.173.3%estimated ± 2.7 pp, medium confidence
149Kimi K273.3%estimated ± 2.7 pp, medium confidence
150GPT-5.4 nano73.0%estimated ± 3.0 pp, low confidence
151A.X K272.8%estimated ± 4.7 pp, medium confidence
152o1-preview72.6%estimated ± 3.2 pp, high confidence
153Grok 472.6%measured
154Gemini 2.5 Pro72.4%estimated ± 2.7 pp, medium confidence
155Nemotron 3.5 Bolt 74B-A9B72.3%estimated ± 4.7 pp, medium confidence
156GPT-5.172.1%estimated ± 3.0 pp, low confidence
157DeepSeek V3 032472.0%estimated ± 4.7 pp, medium confidence
158North Mini Code71.9%estimated ± 4.7 pp, medium confidence
159Spark-X2.6-4B71.9%estimated ± 2.7 pp, medium confidence
160Gemma 4 12B71.8%estimated ± 3.2 pp, high confidence
161Mercury 2.571.8%estimated ± 4.7 pp, medium confidence
162Claude 3.7 Sonnet71.7%estimated ± 2.7 pp, medium confidence
163GPT-OSS 120B71.6%measured
164DeepSeek V3.271.5%measured
165MiniMax M2.771.4%measured
166Command A+71.1%estimated ± 3.2 pp, high confidence
167Ternary Bonsai 2 27B71.0%estimated ± 2.7 pp, medium confidence
168GPT-OSS 20B71.0%measured
169Nemotron 3 Super 100B70.9%estimated ± 2.7 pp, low confidence
170Mistral Small 470.8%estimated ± 3.2 pp, high confidence
171Claude Sonnet 4.5 Thinking70.7%estimated ± 3.0 pp, low confidence
172Trinity-Large-Preview70.5%estimated ± 3.2 pp, high confidence
173Trinity-Large-Thinking70.5%estimated ± 3.2 pp, high confidence
174Ling 2.6 Flash70.4%estimated ± 3.2 pp, high confidence
175GPT-5.1-Codex-Max70.3%estimated ± 3.0 pp, low confidence
176GLM-4.7-Flash70.3%estimated ± 2.7 pp, low confidence
177Gemini 1.5 Pro70.0%estimated ± 3.2 pp, high confidence
178DeepSeek-R1-052869.6%estimated ± 2.7 pp, low confidence
179GPT-4 Turbo69.5%estimated ± 3.2 pp, high confidence
180Granite 4.2 30B69.3%estimated ± 2.7 pp, low confidence
181GPT-4.1 mini69.1%estimated ± 3.2 pp, medium confidence
182Mistral Large 369.1%estimated ± 3.2 pp, medium confidence
183Claude Opus 4.5 Thinking69.0%estimated ± 3.0 pp, low confidence
184Claude 3 Opus69.0%estimated ± 3.2 pp, medium confidence
185MiniMax M1 80k68.8%estimated ± 2.7 pp, low confidence
186Gemini 3 Flash68.7%estimated ± 3.0 pp, low confidence
187GPT-568.6%estimated ± 3.0 pp, low confidence
188GPT-4.168.2%estimated ± 2.7 pp, low confidence
189Llama 4 Maverick68.1%estimated ± 3.2 pp, medium confidence
190Celeris-167.7%estimated ± 3.2 pp, medium confidence
191Nemotron 3 Nano 30B67.7%estimated ± 3.2 pp, medium confidence
192ZAYA1-74B-Preview67.6%estimated ± 2.7 pp, low confidence
193Nemotron 3 Nano Omni 30B A3B67.5%estimated ± 3.2 pp, medium confidence
194Nemotron 3.5 Lightning 30B A3B NVFP467.4%estimated ± 2.7 pp, low confidence
195Mistral Small 3.167.2%estimated ± 4.7 pp, low confidence
196Ultravox v0.6 Llama 3.3 70B67.0%estimated ± 3.2 pp, medium confidence
197GPT-4o mini66.9%estimated ± 3.2 pp, medium confidence
198GPT-4.1 nano66.8%estimated ± 3.2 pp, medium confidence
199LFM2.5-24B-A2B66.6%estimated ± 2.7 pp, low confidence
200Gemma 3 27B66.6%estimated ± 3.2 pp, medium confidence
201Gemma 4 E4B66.4%estimated ± 3.2 pp, medium confidence
202Solar Pro 366.3%estimated ± 4.7 pp, low confidence
203Granite 4.2 3B66.2%estimated ± 4.7 pp, low confidence
204Llama 4 Scout66.1%estimated ± 3.2 pp, medium confidence
205LFM2.5-2.6B66.0%estimated ± 3.2 pp, medium confidence
206K-Exaone65.9%estimated ± 2.7 pp, low confidence
207Gemma 4 E2B65.8%estimated ± 3.2 pp, medium confidence
208o3-mini65.8%estimated ± 2.7 pp, low confidence
209LLaDA2.2-flash65.8%estimated ± 2.7 pp, low confidence
210DeepSeek-R165.8%estimated ± 2.7 pp, low confidence
211Ling 3.0 Tiny65.7%estimated ± 4.7 pp, low confidence
212Claude 3.5 Sonnet65.7%estimated ± 2.7 pp, low confidence
213Granite 4.2 8B65.1%estimated ± 2.7 pp, low confidence
214Mellum2.1-12B-A2.5B-Thinking64.8%estimated ± 2.7 pp, low confidence
215Devstral Small64.7%estimated ± 2.7 pp, low confidence
216MiniCPM5-2B64.5%estimated ± 2.7 pp, low confidence
217Sarvam 105B63.9%estimated ± 2.7 pp, low confidence
218Qwen3.5 Plus63.9%estimated ± 3.0 pp, low confidence
219LFM3-1.7B63.2%estimated ± 2.7 pp, low confidence
220MiniMax M2.562.7%estimated ± 3.0 pp, low confidence
221DeepSeek V362.5%estimated ± 2.7 pp, low confidence
222Claude 3.5 Haiku61.9%estimated ± 2.7 pp, low confidence
223Gemini 3 Pro61.9%estimated ± 3.0 pp, low confidence
224GPT-5 mini61.7%estimated ± 3.0 pp, low confidence
225GPT-5.1-Codex60.1%estimated ± 3.0 pp, low confidence
226Sarvam 30B58.9%estimated ± 2.7 pp, low confidence
227Claude Haiku 4.5 Thinking57.0%estimated ± 3.0 pp, low confidence
228Qwen3 Max30.4%estimated ± 3.0 pp, low confidence
229GLM-4.627.9%estimated ± 3.0 pp, low confidence
230Grok 4.1 Fast (Reasoning)13.2%estimated ± 3.0 pp, low confidence
231Gemini 3.1 Flash-Lite0.0%estimated ± 3.0 pp, low confidence
232Grok 4 Fast (Reasoning)0.0%estimated ± 3.0 pp, low confidence

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General