Independent model intelligence

Find the right AI model,
not just the loudest one.

Compare current model capability, context, modalities, public preference and API economics in one continuously refreshed decision surface.

LiveBench 2.0 LMArena models.dev OpenRouter AIHOT cross-check
Ranked now159of 231 tracked models
Public evidence536verified benchmark results
Evaluation map96reviewed upstream sources
Builders / multimodal20 / 137providers · rich-input models
Leaders right now

Knowledge & instruction

Direct public data · prices never affect capability rank
01
MetaMuse Wonder 1.0
Knowledge & instruction88.0
1M context$20 / 1M out
02
AnthropicClaude Haiku 5.5
Knowledge & instruction81.9
1M context$0.56 / 1M out
03
MetaMuse Spark Open
Knowledge & instruction80.3
1M context$4.25 / 1M out
52 matching models · select up to four to compare
Knowledge & instruction. A score of 50 is the current reference-set average. Price and speed are shown for comparison but never affect rank.
Full ranking

Knowledge & instruction

Open a model for verified evidence, specifications and related releases
RankModelCapabilityContextAPI pricingEvidenceCompare
02—Muse Wonder 1.0Meta · museMultimodalReasoningTools88.0Knowledge & instruction1MLaunched 2026-10-12$4 in$20 out4 public results1 source · 90% coverage
18—Claude Haiku 5.5Anthropic · claude-haikuMultimodalReasoningTools81.9Knowledge & instruction1MLaunched 2026-10-07$0.11 in$0.56 out8 public results3 sources · 100% coverage
38—Muse Spark OpenMeta · museMultimodalReasoningTools80.3Knowledge & instruction1MLaunched 2026-10-15$1.25 in$4.25 out4 public results1 source · 90% coverage
05—GPT-6 AstraOpenAI · gpt-astraMultimodalReasoningTools79.8Knowledge & instruction1.1MLaunched 2026-09-04$12.27 in$61.36 out9 public results3 sources · 100% coverage
06—Mistral Large 4 0Mistral AIMultimodalReasoningTools79.7Knowledge & instruction1MLaunched 2026-10-06$0.7 in$2.1 out4 public results1 source · 90% coverage
09—Gemini 4 ArgonGoogle · geminiMultimodalReasoningTools78.1Knowledge & instruction1MLaunched 2026-10-13$2.25 in$11.25 out9 public results3 sources · 100% coverage
07—GPT-6.1 SolOpenAI · gpt-solMultimodalReasoningTools77.9Knowledge & instruction1.1MLaunched 2026-09-29$2.4 in$12 out9 public results3 sources · 100% coverage
03—Claude Opus 5.5Anthropic · claude-opusMultimodalReasoningTools77.7Knowledge & instruction1MLaunched 2026-09-22$4.4 in$22 out9 public results3 sources · 100% coverage
20—MiMo-V2.6-ProXiaomi · mimoMultimodalReasoningTools76.7Knowledge & instruction1.1MLaunched 2026-09-22$0.44 in$0.87 out8 public results3 sources · 100% coverage
04—Claude Fable 5.1Anthropic · claude-fableMultimodalReasoningTools76.4Knowledge & instruction1MLaunched 2026-09-01$10 in$50 out9 public results3 sources · 100% coverage
10—Qwen3.8 FlashAlibaba Qwen · qwenMultimodalReasoningTools74.6Knowledge & instruction1MLaunched 2026-08-26$0.16 in$0.47 out4 public results1 source · 90% coverage
26—Grok 4.7xAI · grokMultimodalReasoningTools74.6Knowledge & instruction500kLaunched 2026-09-21$2 in$6 out9 public results3 sources · 100% coverage
15—Kimi K3.1 ProMoonshot AI · kimi-k3MultimodalReasoningTools74.5Knowledge & instruction1MLaunched 2026-10-23$2 in$10 out8 public results2 sources · 100% coverage
08—Claude Sonnet 5.5Anthropic · claude-sonnetMultimodalReasoningTools74.1Knowledge & instruction1MLaunched 2026-09-28$2.25 in$11.25 out9 public results3 sources · 100% coverage
16—Gemini 3.8 FlashGoogle · gemini-flashMultimodalReasoningTools73.8Knowledge & instruction1MLaunched 2026-09-02$0.83 in$4.13 out9 public results3 sources · 100% coverage
13—deepseek-v4.1-flash-maxDeepSeek · deepseek-flashMultimodalReasoningTools73.5Knowledge & instruction1MLaunched 2026-09-10$0.15 in$0.6 out8 public results3 sources · 100% coverage
23—Grok 4.6xAI · grokMultimodalReasoningTools73.4Knowledge & instruction500kLaunched 2026-08-12$2 in$6 out9 public results3 sources · 100% coverage
37—Kimi K3Moonshot AI · kimi-k3MultimodalReasoningTools73.0Knowledge & instruction1MLaunched 2026-07-16$3 in$15 out9 public results3 sources · 100% coverage
——Step 5 PreviewstepfunMultimodalReasoningTools72.8Knowledge & instruction1MLaunched 2026-09-16$1 in$2.7 out0 public results0 sources · 0% coverage
17—Step 5 PreviewIndependentMultimodalReasoningTools72.8Knowledge & instruction1MLaunched 2026-09-16$1 in$2.7 out8 public results3 sources · 100% coverage
28—Beam 1.0Independent · beamReasoningToolsStructured72.8Knowledge & instruction1MLaunched 2026-10-05$0 in$0 out4 public results1 source · 90% coverage
12—Minimax M3.1 FlashMiniMax · minimaxMultimodalReasoningTools72.7Knowledge & instruction1MLaunched 2026-10-19$0.15 in$0.6 out8 public results2 sources · 100% coverage
14—Deepseek V4.1 Flash 1017DeepSeek · deepseek-flashMultimodalReasoningTools72.7Knowledge & instruction1MLaunched 2026-10-17$0.2 in$0.61 out8 public results2 sources · 100% coverage
11—GPT-6.1 TerraOpenAI · gpt-terraMultimodalReasoningTools70.9Knowledge & instruction1.1MLaunched 2026-10-26$1.18 in$5.88 out8 public results2 sources · 100% coverage
60—Nemotron Super Vl 3.5 Nvfp4NVIDIAText70.9Knowledge & instruction—Launch date pending— in— out4 public results1 source · 90% coverage
19—GLM-5.3-FlashZhipu AI · glm-flashMultimodalReasoningTools70.1Knowledge & instruction1MLaunched 2026-08-26$0.1 in$0.35 out5 public results2 sources · 95% coverage
31—Claude Fable 5Anthropic · claude-fableMultimodalReasoningTools69.9Knowledge & instruction1MLaunched 2026-06-09$10 in$50 out9 public results3 sources · 100% coverage
62—Deepseek V4 Flash VisionDeepSeek · deepseek-flashMultimodalReasoningTools69.2Knowledge & instruction1MLaunched 2026-08-21$0.22 in$0.65 out4 public results1 source · 90% coverage
30—Claude Opus 5Anthropic · claude-opusMultimodalReasoningTools69.0Knowledge & instruction1MLaunched 2026-07-24$5.5 in$27.5 out9 public results3 sources · 100% coverage
29—Muse Wonder FullMeta · museMultimodalReasoningTools68.1Knowledge & instruction1MLaunched 2026-10-12$4 in$20 out4 public results1 source · 90% coverage
39—GPT-5.6 SolOpenAI · gpt-solMultimodalReasoningTools67.8Knowledge & instruction1.1MLaunched 2026-07-09$4.4 in$22 out9 public results3 sources · 100% coverage
46—Smaug Nemotron 3.5 Nvfp4NVIDIAText67.8Knowledge & instruction—Launch date pending— in— out4 public results1 source · 90% coverage
48—Nemotron Super Vl 3.5NVIDIA · nemotronText66.7Knowledge & instruction—Launched 2026-03-11— in— out4 public results1 source · 90% coverage
25—Muse Spark 1.3Meta · museMultimodalReasoningTools66.6Knowledge & instruction1MLaunched 2026-09-02$1.25 in$4.25 out9 public results3 sources · 100% coverage
52—Gemini 3.7 FlashGoogle · gemini-flashMultimodalReasoningTools66.3Knowledge & instruction1MLaunched 2026-08-13$0.75 in$3.75 out9 public results3 sources · 100% coverage
45—Qwen3.8-MaxAlibaba Qwen · qwenMultimodalReasoningTools65.9Knowledge & instruction1MLaunched 2026-08-03$2 in$6 out8 public results3 sources · 100% coverage
35—Muse WonderMeta · museMultimodalReasoningTools65.4Knowledge & instruction1MLaunched 2026-10-12$4 in$20 out5 public results2 sources · 95% coverage
40—GPT-5.5OpenAI · gptMultimodalReasoningTools65.4Knowledge & instruction1MLaunched 2026-04-23$5 in$30 out9 public results3 sources · 100% coverage
42—GPT-6 SolOpenAI · gpt-solMultimodalReasoningTools65.1Knowledge & instruction1.1MLaunched 2026-09-22$2.4 in$12 out9 public results3 sources · 100% coverage
53—Claude Opus 4.8Anthropic · claude-opusMultimodalReasoningTools64.4Knowledge & instruction1MLaunched 2026-05-28$5.44 in$27.19 out9 public results3 sources · 100% coverage
——DeepSeek-V4.1-Flash-20260910DeepSeek · deepseek-flashMultimodalReasoningTools62.0Knowledge & instruction1MLaunched 2026-09-10$0.3 in$1.2 out0 public results0 sources · 0% coverage
——DeepSeek V4.1 FlashDeepSeek · deepseek-flashMultimodalReasoningTools62.0Knowledge & instruction1MLaunched 2026-09-10$0.6 in$2.4 out2 public results1 source · 45% coverage
51—GLM-5.3Zhipu AI · glmReasoningToolsStructured61.4Knowledge & instruction1MLaunched 2026-08-14$0.51 in$2.05 out9 public results3 sources · 100% coverage
54—Inkling 2 SmallThinking Machines · lingMultimodalReasoningTools60.9Knowledge & instruction262.1kLaunched 2026-07-15$3.74 in$9.36 out4 public results1 source · 90% coverage
27—GPT-5.6 TerraOpenAI · gpt-terraMultimodalReasoningTools60.4Knowledge & instruction1.1MLaunched 2026-07-09$2.2 in$13.2 out9 public results3 sources · 100% coverage
55—InklingThinking Machines · lingMultimodalReasoningTools59.9Knowledge & instruction262.1kLaunched 2026-07-15$3.74 in$9.36 out5 public results2 sources · 95% coverage
21—Qwen3.8-27BAlibaba Qwen · qwenMultimodalReasoningTools59.4Knowledge & instruction1MLaunched 2026-08-14$0.1 in$0.4 out5 public results2 sources · 95% coverage
——Gemma 4 31BGoogle · gemmaMultimodalReasoningTools44.4Knowledge & instruction262.1kLaunched 2026-04-02$0.1 in$0.34 out5 public results2 sources · 95% coverage
22—GPT-6 LunaOpenAI · gpt-lunaMultimodalReasoningTools43.5Knowledge & instruction1.1MLaunched 2026-09-22$0.12 in$0.6 out5 public results2 sources · 95% coverage
——Nemotron Ultra 3NVIDIA · nemotronReasoningToolsStructured43.2Knowledge & instruction202.8kLaunched 2026-06-04$0.6 in$2.4 out4 public results1 source · 90% coverage
——Nemotron Lightning 3.5NVIDIA · nemotronReasoningToolsStructured40.0Knowledge & instruction262.1kLaunched 2026-08-15$0.05 in$0.2 out4 public results1 source · 90% coverage
81—Mistral Large 4Mistral AI · mistral-largeMultimodalReasoningTools28.1Knowledge & instruction1MLaunched 2026-10-06$0.77 in$2.35 out5 public results2 sources · 95% coverage