Model Explorer
Weights
Organization
Total parameters
Runs on my hardware
Filters open models whose Q4 quant fits in VRAM/unified memory.
Capabilities
568 models
Sort
GPT-6 Astra
CLOSEDOpenAI · Sep 2026
—1.05M ctx$10/$50
3305.4
ReasoningVisionFunction callingTool useAgentic
Muse Spark 1.3
CLOSEDMeta · Sep 2026
—1M ctx$1.25/$4.25
3269.8
ReasoningVisionFunction callingTool useAgentic
Claude Fable 5.1
CLOSEDAnthropic · Sep 2026
—1M ctx$10/$50
3144.2
ReasoningVisionFunction callingTool useAgentic
Qwen3.8-Max-0902
CLOSEDAlibaba · Sep 2026
2400B·95Ba1M ctx$2/$6
3092.0
ReasoningVisionFunction callingTool useAgentic
Claude Opus 5
CLOSEDAnthropic · Jul 2026
—1M ctx$5/$25
3078.5
ReasoningVisionFunction callingTool useAgentic
GLM-5.3
OPENZhipu AI / Z.ai (Tsinghua) · Aug 2026
744B·40Ba200K ctx$1.4/$4.4
3029.9
ReasoningFunction callingTool useAgentic
GLM-5.3-Flash
OPENZhipu AI / Z.ai (Tsinghua) · Aug 2026
320B·18Ba200K ctx$0.14/$0.44
3017.6
ReasoningVisionFunction callingTool useAgentic
Kimi K3
OPENMoonshot AI · Jul 2026
2800B1.024M ctx$3/$15
3005.1
ReasoningVisionFunction callingTool useAgentic
Grok 4.6
CLOSEDSpaceXAI · Aug 2026
—500K ctx$2/$6
2960.9
ReasoningVisionFunction callingTool useAgentic
Gemini 3.8 Flash
CLOSEDGoogle · Sep 2026
—1M ctx$0.75/$3.75
2949.4
ReasoningVisionFunction callingTool useAgentic
Claude Fable 5
CLOSEDAnthropic · Jun 2026
—1M ctx$10/$50
2928.6
ReasoningVisionFunction callingTool useAgentic
GPT-5.6 Sol
CLOSEDOpenAI · Jul 2026
—1.05M ctx$5/$30
2915.3
ReasoningVisionFunction callingTool useAgentic
Qwen3.8 Max
CLOSEDAlibaba · Aug 2026
2400B·95Ba1M ctx$2/$6
2863.4
ReasoningVisionFunction callingTool useAgentic
GPT-5.4 Pro
CLOSEDOpenAI · Mar 2026
—1.05M ctx$30/$180
2759.5
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.8
CLOSEDAnthropic · May 2026
—1M ctx$5/$25
2756.0
ReasoningVisionFunction callingTool useAgentic
Gemini 3.7 Flash
CLOSEDGoogle · Aug 2026
—1M ctx$0.75/$3.75
2710.0
ReasoningVisionFunction callingTool useAgentic
GPT-5.6 Terra
CLOSEDOpenAI · Jul 2026
—1.05M ctx$2.5/$15
2688.0
ReasoningVisionFunction callingTool useAgentic
GPT-5.5 (xHigh)
CLOSEDOpenAI · Apr 2026
—1.05M ctx$5/$30
2677.5
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.7
CLOSEDAnthropic · Apr 2026
—1M ctx$5/$25
2674.5
ReasoningVisionFunction callingTool useAgentic
Muse Spark 1.2
CLOSEDMeta · Aug 2026
—1M ctx$1.25/$4.25
2670.8
ReasoningVisionFunction callingTool useAgentic
Qwen3.8-Flash-Next
OPENAlibaba · Aug 2026
125B·6Ba262K ctx$0.15/$0.47
2659.1
ReasoningVisionFunction callingTool useAgentic
GPT-5.6 Luna
CLOSEDOpenAI · Jul 2026
—1.05M ctx$1/$6
2651.5
ReasoningVisionFunction callingTool useAgentic
GPT-5.5
CLOSEDOpenAI · Apr 2026
—1.05M ctx$5/$30
2634.5
ReasoningVisionFunction callingTool useAgentic
Seed 2.1 Pro
CLOSEDByteDance · Jun 2026
—256K ctx$0.85/$4.15
2617.1
ReasoningVisionFunction callingTool useAgentic
Muse Spark 1.1 (xhigh)
CLOSEDMeta · Jul 2026
—1M ctx$1.25/$4.25
2563.9
ReasoningVisionFunction callingTool useAgentic
Gemini 3.1 Pro
CLOSEDGoogle · Feb 2026
—1M ctx$2/$12
2561.9
ReasoningVisionFunction callingTool useAgentic
GPT-5.5 (High)
CLOSEDOpenAI · Apr 2026
—1.05M ctx$5/$30
2560.2
ReasoningVisionFunction callingTool useAgentic
Gemini 3.6 Flash
CLOSEDGoogle · Jul 2026
—1M ctx$1.5/$7.5
2551.9
ReasoningVisionFunction callingTool useAgentic
Qwen3.7-Max
CLOSEDAlibaba · May 2026
—1M ctx$2.5/$7.5
2506.5
ReasoningFunction callingTool useAgentic
Qwen3.8-27B
OPENAlibaba · Aug 2026
27B262K ctxweights
2506.1
ReasoningVisionFunction callingTool useAgentic
Gemini 3.5 Flash
CLOSEDGoogle · May 2026
—1M ctx$1.5/$9
2498.9
ReasoningVisionFunction callingTool useAgentic
MiniMax M3
OPENMiniMax · Jun 2026
428B·23Ba1.024M ctx$0.3/$1.2
2488.4
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V4-Pro (Think Max)
OPENDeepSeek · Apr 2026
1600B·49Ba1M ctx$0.435/$0.87
2487.3
ReasoningFunction callingTool useAgentic
Kimi K2.6
OPENMoonshot AI · Apr 2026
1000B·32Ba256K ctx$0.95/$4
2471.3
ReasoningVisionFunction callingTool useAgentic
GPT-5.2
CLOSEDOpenAI · Dec 2025
—400K ctx$1.75/$14
2469.9
ReasoningVisionFunction callingTool useAgentic
Hy3
OPENTencent · Jul 2026
295B·21Ba256K ctx$0.14/$0.55
2460.0
ReasoningFunction callingTool useAgentic
GPT-5.4
CLOSEDOpenAI · Mar 2026
—1.05M ctx$2.5/$15
2445.3
ReasoningVisionFunction callingTool useAgentic
Grok 4.5
CLOSEDSpaceXAI · Jul 2026
—500K ctx$2/$6
2444.0
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.6
CLOSEDAnthropic · Feb 2026
—1M ctx$5/$25
2433.7
ReasoningVisionFunction callingTool useAgentic
GLM-5.2 (Max)
OPENZhipu AI / Z.ai (Tsinghua) · Jun 2026
744B·40Ba1M ctx$1.4/$4.4
2425.7
ReasoningFunction callingTool useAgentic
Claude Sonnet 5
CLOSEDAnthropic · Jun 2026
—1M ctx$3/$15
2420.4
ReasoningVisionFunction callingTool useAgentic
Claude Sonnet 4.6
CLOSEDAnthropic · Feb 2026
—1M ctx$3/$15
2417.2
ReasoningVisionFunction callingTool useAgentic
Qwen3.7-Plus
CLOSEDAlibaba · Jun 2026
—1M ctx$0.4/$1.6
2387.6
ReasoningVisionFunction callingTool useAgentic
Inkling-Small
OPENThinking Machines Lab · Jul 2026
276B·12Ba1M ctx$0.58/$1.44
2384.1
ReasoningVisionFunction callingTool useAgentic
GPT-5 Pro
CLOSEDOpenAI · Aug 2025
—400K ctx$15/$120
2355.3
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V4-Pro (Think High)
OPENDeepSeek · Apr 2026
1600B·49Ba1M ctx$0.435/$0.87
2351.3
ReasoningFunction callingTool useAgentic
Gemini 3 Pro
CLOSEDGoogle · Nov 2025
—1M ctx$2/$12
2346.0
ReasoningVisionFunction callingTool useAgentic
Gemini 3 Flash
CLOSEDGoogle · Dec 2025
—1M ctx$0.5/$3
2326.6
ReasoningVisionFunction callingTool useAgentic
Grok 4 Heavy
CLOSEDSpaceXAI · Jul 2025
—256K ctx—
2302.0
ReasoningVisionFunction callingTool useAgentic
Qwen3.6-Plus
CLOSEDAlibaba · Apr 2026
—1M ctx$0.5/$3
2290.1
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V4-Flash (Think Max)
OPENDeepSeek · Apr 2026
284B·13Ba1M ctx$0.14/$0.28
2289.9
ReasoningFunction callingTool useAgentic
Grok 4.3 (High)
CLOSEDSpaceXAI · Apr 2026
—1M ctx$1.25/$2.5
2270.6
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.5 (High)
CLOSEDAnthropic · Nov 2025
—200K ctx$5/$25
2261.5
ReasoningVisionFunction callingTool useAgentic
Kimi K2.5
OPENMoonshot AI · Jan 2026
1000B·32Ba256K ctx$0.6/$3
2233.3
ReasoningVisionFunction callingTool useAgentic
Inkling
OPENThinking Machines Lab · Jul 2026
975B·41Ba1M ctx$1.87/$4.68
2231.4
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.5
CLOSEDAnthropic · Nov 2025
—200K ctx$5/$25
2228.2
ReasoningVisionFunction callingTool useAgentic
GLM-5.1
OPENZhipu AI / Z.ai (Tsinghua) · Apr 2026
744B·40Ba200K ctx$1.4/$4.4
2224.1
ReasoningFunction callingTool useAgentic
Kimi K2.7 Code
OPENMoonshot AI · Jun 2026
1000B·32Ba256K ctx$0.95/$4
2223.5
ReasoningVisionFunction callingTool useAgentic
Grok Build 0.1
CLOSEDSpaceXAI · May 2026
—256K ctx$1/$2
2223.1
ReasoningVisionFunction callingTool useAgentic
GPT-5.1
CLOSEDOpenAI · Nov 2025
—400K ctx$1.25/$10
2215.2
ReasoningVisionFunction callingTool useAgentic
Step 3.7 Flash
OPENStepFun · May 2026
198B·11Ba262K ctx$0.2/$1.15
2212.7
VisionFunction callingTool useAgentic
DeepSeek-V4-Flash (Think High)
OPENDeepSeek · Apr 2026
284B·13Ba1M ctx$0.14/$0.28
2202.0
ReasoningFunction callingTool useAgentic
MiMo-V2.5-Pro
OPENXiaomi · Apr 2026
1020B·42Ba1M ctx$0.435/$0.87
2187.4
ReasoningFunction callingTool useAgentic
Kimi K2 Thinking
OPENMoonshot AI · Nov 2025
1000B·32Ba256K ctx$0.6/$2.5
2161.6
ReasoningFunction callingTool useAgentic
Nemotron 3 Ultra 550B A55B
OPENNVIDIA · Jun 2026
550B·55Ba1M ctx$0.5/$2.2
2155.8
ReasoningFunction callingTool useAgentic
Qwen3.5-397B-A17B
OPENAlibaba · Feb 2026
397B·17Ba256K ctxweights
2146.0
ReasoningVisionFunction callingTool useAgentic
MiniMax M2.7
OPENMiniMax · Mar 2026
230B·10Ba200K ctx$0.3/$1.2
2123.7
ReasoningFunction callingTool useAgentic
Gemini 3.5 Flash-Lite
CLOSEDGoogle · Jul 2026
—1M ctx$0.3/$2.5
2105.3
ReasoningVisionFunction callingTool useAgentic
GLM-5
OPENZhipu AI / Z.ai (Tsinghua) · Feb 2026
744B·40Ba200K ctx$1/$3.2
2096.8
ReasoningFunction callingTool useAgentic
Muse Glimmer 30B
OPENMeta · Aug 2026
29.6B131K ctxweights
2077.8
ReasoningVisionFunction callingTool useAgentic
GPT-5
CLOSEDOpenAI · Aug 2025
—400K ctx$1.25/$10
2074.5
ReasoningVisionFunction callingTool useAgentic
OpenAI o3-pro
CLOSEDOpenAI · Jun 2025
—200K ctx$20/$80
2067.0
ReasoningVisionFunction callingTool useAgentic
GPT-5.4 mini
CLOSEDOpenAI · Mar 2026
—400K ctx$0.75/$4.5
2065.9
ReasoningVisionFunction callingTool useAgentic
openPangu-2.0-Flash (Thinking)
OPENHuawei · Jun 2026
92B·6Ba512K ctxweights
2057.2
ReasoningFunction callingTool useAgentic
MiMo-V2.5
OPENXiaomi · Apr 2026
310B·15Ba1M ctx$0.14/$0.28
2050.2
ReasoningVisionFunction callingTool useAgentic
Qwen3.6-27B
OPENAlibaba · Apr 2026
27B256K ctx$0.6/$3.6
2027.4
ReasoningVisionFunction callingTool useAgentic
Qwen3.5-122B-A10B
OPENAlibaba · Feb 2026
122B·10Ba256K ctxweights
2004.1
ReasoningVisionFunction callingTool useAgentic
Gemma 4
OPENGoogle · Apr 2026
30.7B256K ctxweights
1976.5
ReasoningVisionFunction callingTool useAgentic
Qwen3.6-35B-A3B
OPENAlibaba · Apr 2026
35B·3Ba256K ctxweights
1973.2
ReasoningVisionFunction callingTool useAgentic
Claude Sonnet 4.5
CLOSEDAnthropic · Sep 2025
—200K ctx$3/$15
1938.8
ReasoningVisionFunction callingTool useAgentic
Grok-4
CLOSEDSpaceXAI · Jul 2025
—256K ctx$3/$15
1937.8
ReasoningVisionFunction callingTool useAgentic
GPT-5.4 nano
CLOSEDOpenAI · Mar 2026
—400K ctx$0.2/$1.25
1929.3
ReasoningVisionFunction callingTool useAgentic
GLM-4.6
OPENZhipu AI / Z.ai (Tsinghua) · Sep 2025
357B·32Ba200K ctx$0.6/$2.2
1899.3
ReasoningFunction callingTool useAgentic
OpenAI o3
CLOSEDOpenAI · Apr 2025
—200K ctx$2/$8
1896.3
ReasoningVisionFunction callingTool useAgentic
Gemini 3.1 Flash-Lite
CLOSEDGoogle · Mar 2026
—1M ctx$0.25/$1.5
1896.0
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V3.1 (Thinking)
OPENDeepSeek · Aug 2025
671B·37Ba128K ctx$0.56/$1.68
1890.4
ReasoningFunction callingTool useAgentic
Grok-3 (Think)
CLOSEDSpaceXAI · Feb 2025
—131K ctx$3/$15
1887.6
ReasoningFunction callingTool useAgentic
Trinity-Large-Thinking
OPENArcee AI · Apr 2026
398B·13Ba512K ctx$0.25/$0.8
1879.8
ReasoningFunction callingTool useAgentic
Qwen3.5-Plus
CLOSEDAlibaba · Feb 2026
—1M ctx$0.4/$2.4
1864.3
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V3.2-Exp
OPENDeepSeek · Sep 2025
685B·37Ba128K ctx$0.28/$0.42
1853.8
ReasoningFunction callingTool useAgentic
OpenAI o4-mini
CLOSEDOpenAI · Apr 2025
—200K ctx$1.1/$4.4
1852.4
ReasoningVisionFunction callingTool useAgentic
GPT-5 mini
CLOSEDOpenAI · Aug 2025
—400K ctx$0.25/$2
1846.1
ReasoningVisionFunction callingTool useAgentic
gpt-oss-120b (High)
OPENOpenAI · Aug 2025
117B·5.1Ba128K ctxweights
1839.3
ReasoningFunction callingTool useAgentic
Gemini 2.5 Pro
CLOSEDGoogle · Mar 2025
—1M ctx$1.25/$10
1839.1
ReasoningVisionFunction callingTool useAgentic
EXAONE 4.5
OPENLG AI Research · Apr 2026
33B256K ctxweights
1831.7
ReasoningVisionFunction calling
Qwen3-235B-A22B-Thinking-2507
OPENAlibaba · Jul 2025
235B·22Ba256K ctx$0.23/$2.3
1827.4
ReasoningFunction callingTool useAgentic
MiniMax-M2
OPENMiniMax · Oct 2025
229.9B·9.8Ba192K ctx$0.3/$1.2
1824.9
ReasoningFunction callingTool useAgentic
Qwen3.5-Omni-Plus
CLOSEDAlibaba · Mar 2026
—256K ctx$1.4/$8.3
1817.9
ReasoningVisionFunction callingTool useAgentic
Nemotron 3 Super
OPENNVIDIA · Mar 2026
120B·12Ba1M ctxweights
1805.3
ReasoningFunction callingTool useAgentic
Gemma 4 26B A4B
OPENGoogle · Apr 2026
25.2B·3.8Ba256K ctxweights
1792.3
ReasoningVisionFunction callingTool useAgentic
GLM-4.5 (355B-A32B)
OPENZhipu AI / Z.ai (Tsinghua) · Jul 2025
355B·32Ba128K ctx$0.6/$2.2
1749.9
ReasoningFunction callingTool useAgentic
DeepSeek-R1-0528
OPENDeepSeek · May 2025
671B·37Ba128K ctx$0.55/$2.19
1741.7
ReasoningFunction callingTool useAgentic
Command A+
OPENCohere · May 2026
218B·25Ba128K ctxweights
1735.1
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4.1
CLOSEDAnthropic · Aug 2025
—200K ctx$15/$75
1733.8
ReasoningVisionFunction callingTool useAgentic
ERNIE 5.0 (Preview)
CLOSEDBaidu · Nov 2025
2400B128K ctx—
1729.3
ReasoningVisionFunction callingTool useAgentic
INTELLECT-3
OPENPrime Intellect · Nov 2025
106B·12Ba98K ctxweights
1719.0
ReasoningFunction callingTool useAgentic
Grok-3 mini
CLOSEDSpaceXAI · Feb 2025
—131K ctx$0.3/$0.5
1697.4
ReasoningFunction callingTool useAgentic
Doubao-Seed-1.6
CLOSEDByteDance · Jun 2025
230B·23Ba256K ctx$0.111/$0.278
1680.5
ReasoningVisionFunction callingTool useAgentic
Claude Opus 4
CLOSEDAnthropic · May 2025
—200K ctx$15/$75
1674.4
ReasoningVisionFunction callingTool useAgentic
Sarvam-105B
OPENSarvam AI · Feb 2026
105B·10.3Ba128K ctxweights
1663.5
ReasoningFunction callingTool useAgentic
DeepSeek-V4-Pro (Non-Think)
OPENDeepSeek · Apr 2026
1600B·49Ba1M ctx$0.435/$0.87
1643.5
ReasoningFunction callingTool useAgentic
Qwen3-Max
CLOSEDAlibaba · Sep 2025
—256K ctx$1.2/$6
1635.1
ReasoningFunction callingTool useAgentic
OpenAI o3-mini (High)
CLOSEDOpenAI · Jan 2025
—200K ctx$1.1/$4.4
1622.7
ReasoningFunction callingTool useAgentic
GLM-4.5-Air (106B-A12B)
OPENZhipu AI / Z.ai (Tsinghua) · Jul 2025
106B·12Ba128K ctx$0.2/$1.1
1613.7
ReasoningFunction callingTool useAgentic
Hunyuan-T1
CLOSEDTencent · Mar 2025
—256K ctx$0.15/$0.59
1610.8
ReasoningFunction callingTool use
Step-3
OPENStepFun · Jul 2025
321B·38Ba64K ctx$0.57/$1.42
1604.5
ReasoningVisionFunction callingTool use
openPangu-2.0-Flash (Non-Thinking)
OPENHuawei · Jun 2026
92B·6Ba512K ctxweights
1603.9
ReasoningFunction callingTool useAgentic
Gemini 2.5 Flash
CLOSEDGoogle · May 2025
—1M ctx$0.3/$2.5
1603.0
ReasoningVisionFunction callingTool useAgentic
Seed1.5-Thinking
CLOSEDByteDance · Apr 2025
200B·20Ba128K ctx$0.556/$2.222
1598.4
ReasoningFunction callingTool useAgentic
DeepSeek-V4-Flash (Non-Think)
OPENDeepSeek · Apr 2026
284B·13Ba1M ctx$0.14/$0.28
1596.6
ReasoningFunction callingTool useAgentic
GPT-5 nano
CLOSEDOpenAI · Aug 2025
—400K ctx$0.05/$0.4
1583.9
ReasoningVisionFunction callingTool useAgentic
Seed-OSS-36B
OPENByteDance · Aug 2025
36B512K ctxweights
1576.9
ReasoningFunction callingTool useAgentic
OpenAI o1
CLOSEDOpenAI · Dec 2024
—200K ctx$15/$60
1567.5
ReasoningVisionFunction callingTool useAgentic
Claude Sonnet 4
CLOSEDAnthropic · May 2025
—200K ctx$3/$15
1566.9
ReasoningVisionFunction callingTool useAgentic
Qwen3-235B-A22B-Instruct-2507
OPENAlibaba · Jul 2025
235B·22Ba256K ctx$0.23/$0.92
1559.3
ReasoningFunction callingTool useAgentic
Claude Haiku 4.5
CLOSEDAnthropic · Oct 2025
—200K ctx$1/$5
1556.0
ReasoningVisionFunction callingTool useAgentic
DeepSeek-R1 671B
OPENDeepSeek · Jan 2025
671B·37Ba128K ctx$0.55/$2.19
1553.6
Reasoning
LG EXAONE 4.0 32B (Reasoning)
OPENLG AI Research · Jul 2025
32B131K ctxweights
1551.7
ReasoningFunction callingTool useAgentic
MiniMax-M1
OPENMiniMax · Jun 2025
456B·45.9Ba1M ctx$0.4/$2.2
1547.8
ReasoningFunction callingTool useAgentic
gpt-oss-120b (Medium)
OPENOpenAI · Aug 2025
117B·5.1Ba128K ctxweights
1533.5
ReasoningFunction callingTool useAgentic
gpt-oss-20b (High)
OPENOpenAI · Aug 2025
21B·3.6Ba128K ctxweights
1527.6
ReasoningFunction callingTool useAgentic
Gemini 2.0 Flash Thinking (Experimental)
CLOSEDGoogle · Jan 2025
—1M ctx—
1521.9
ReasoningVisionTool use
Falcon-H1R 7B
OPENTII · Jan 2026
7.6B256K ctxweights
1519.0
ReasoningFunction callingTool useAgentic
Llama-3.1-Nemotron-Ultra-253B v1 (Reasoning On)
OPENNVIDIA · Apr 2025
253B128K ctx$2/$6
1518.9
ReasoningFunction callingTool useAgentic
DeepSeek-V3.1 (Non-Thinking)
OPENDeepSeek · Aug 2025
671B·37Ba128K ctx$0.56/$1.68
1517.5
Function callingTool useAgentic
Qwen3-235B-A22B (Thinking)
OPENAlibaba · Apr 2025
235B·22Ba128K ctxweights
1516.1
ReasoningFunction callingTool useAgentic
Hunyuan-A13B
OPENTencent · Jun 2025
80B·13Ba256K ctx$0.07/$0.3
1513.6
ReasoningFunction callingTool useAgentic
Kimi K2
OPENMoonshot AI · Jul 2025
1000B·32Ba128K ctx$0.57/$2.3
1491.9
Function callingTool useAgentic
Nemotron 3 Nano Omni
OPENNVIDIA · Apr 2026
30B·3Ba256K ctx$0.075/$0.3
1485.6
ReasoningVisionFunction callingTool useAgentic
Qwen3-32B (Thinking)
OPENAlibaba · Apr 2025
32.8B128K ctx$0.287/$1.147
1484.0
ReasoningFunction callingTool useAgentic
Nemotron Nano 2 (Nemotron-Nano-9B-v2)
OPENNVIDIA · Aug 2025
9B128K ctx$0.04/$0.16
1482.7
ReasoningFunction callingTool useAgentic
Grok-3 (Standard)
CLOSEDSpaceXAI · Feb 2025
—131K ctx$3/$15
1476.3
Function callingTool use
Magistral Medium
CLOSEDMistral AI · Jun 2025
—40K ctx$2/$5
1464.6
ReasoningFunction callingTool useAgentic
OpenAI o3-mini (Medium)
CLOSEDOpenAI · Jan 2025
—200K ctx$1.1/$4.4
1462.9
ReasoningFunction callingTool useAgentic
Qwen3.5-Omni-Flash
CLOSEDAlibaba · Mar 2026
—256K ctx$0.4/$2.2
1454.8
ReasoningVisionFunction callingTool useAgentic
Qwen3-30B-A3B (Thinking)
OPENAlibaba · Apr 2025
30.5B·3.3Ba128K ctx$0.108/$0.431
1452.3
ReasoningFunction callingTool useAgentic
OLMo 3-Think 32B
OPENAllen Institute for AI (Ai2) · Nov 2025
32.23B64K ctx$0.15/$0.5
1448.0
Reasoning
GPT-4.5
CLOSEDOpenAI · Feb 2025
—128K ctx$75/$150
1444.1
VisionFunction callingTool useAgentic
GPT-4.1
CLOSEDOpenAI · Apr 2025
—1M ctx$2/$8
1437.8
VisionFunction callingTool useAgentic
Claude 3.7 Sonnet
CLOSEDAnthropic · Feb 2025
—200K ctx$3/$15
1423.4
ReasoningVisionFunction callingTool useAgentic
Doubao-1.5-Pro
CLOSEDByteDance · Jan 2025
200B·20Ba256K ctx$0.11/$0.275
1423.0
ReasoningFunction callingTool useAgentic
DeepSeek-R1-Zero
OPENDeepSeek · Jan 2025
671B·37Ba128K ctxweights
1419.6
Reasoning
GLM-Z1-32B-0414
OPENZhipu AI · Apr 2025
32B32K ctxweights
1413.3
ReasoningFunction callingTool useAgentic
ERNIE 4.5 300B-A47B
OPENBaidu · Jun 2025
300B·47Ba128K ctx$0.28/$1.1
1399.9
Function callingTool useAgentic
Magistral Small
OPENMistral AI · Jun 2025
24B128K ctx$0.5/$1.5
1399.6
ReasoningFunction callingTool useAgentic
Qwen3-14B (Thinking)
OPENAlibaba · Apr 2025
14.8B128K ctxweights
1382.9
ReasoningFunction callingTool useAgentic
DeepSeek-R1-Distill-Llama-70B
OPENDeepSeek · Jan 2025
70B128K ctxweights
1364.8
Reasoning
Kimi k1.5
CLOSEDMoonshot AI · Jan 2025
—128K ctx—
1361.3
ReasoningVision
OpenAI o1-preview
CLOSEDOpenAI · Sep 2024
—128K ctx$15/$60
1359.7
Reasoning
GLM-4-32B-0414
OPENZhipu AI · Apr 2025
32B32K ctxweights
1358.6
Function callingTool useAgentic
DeepSeek-V3-0324
OPENDeepSeek · Mar 2025
671B·37Ba128K ctx$0.27/$1.1
1346.8
Function callingTool useAgentic
Gemini 2.0 Pro Experimental
CLOSEDGoogle · Feb 2025
—2M ctx—
1344.8
VisionFunction callingTool useAgentic
gpt-oss-20b (Medium)
OPENOpenAI · Aug 2025
21B·3.6Ba128K ctxweights
1340.3
ReasoningFunction callingTool useAgentic
Llama-3.3-Nemotron-Super-49B v1 (Reasoning On)
OPENNVIDIA · Mar 2025
49.9B128K ctx$0.1/$0.5
1333.5
ReasoningFunction callingTool useAgentic
Qwen3-Coder-480B-A35B
OPENAlibaba · Jul 2025
480B·35Ba256K ctx$1.5/$7.5
1326.5
ReasoningFunction callingTool useAgentic
Claude 3.5 Sonnet (New)
CLOSEDAnthropic · Oct 2024
—200K ctx$3/$15
1320.9
VisionFunction callingTool useAgentic
LG EXAONE Deep 32B
OPENLG AI Research · Mar 2025
32B32K ctxweights
1315.7
Reasoning
Gemini 2.0 Flash
CLOSEDGoogle · Dec 2024
—1M ctx$0.1/$0.4
1312.1
VisionFunction callingTool useAgentic
Qwen3-8B (Thinking)
OPENAlibaba · Apr 2025
8.2B128K ctx$0.072/$0.287
1311.4
ReasoningFunction callingTool useAgentic
OpenAI o3-mini (Low)
CLOSEDOpenAI · Jan 2025
—200K ctx$1.1/$4.4
1309.0
ReasoningFunction callingTool useAgentic
gpt-oss-120b (Low)
OPENOpenAI · Aug 2025
117B·5.1Ba128K ctxweights
1291.7
ReasoningFunction callingTool useAgentic
Mistral Large 3
OPENMistral AI · Dec 2025
675B·41Ba256K ctx$0.5/$1.5
1283.1
VisionFunction callingTool useAgentic
InternVL2.5-78B
OPENOpenGVLab (Shanghai AI Lab) · Dec 2024
78B32K ctxweights
1281.8
Vision
Skywork-R1V (38B)
OPENKunlun Tech · Mar 2025
38B128K ctxweights
1280.5
ReasoningVision
Phi-4-reasoning
OPENMicrosoft · Apr 2025
14B32K ctxweights
1268.4
Reasoning
Phi-4-reasoning (14B)
OPENMicrosoft · Apr 2025
14B32K ctxweights
1268.4
ReasoningFunction callingTool use
Hunyuan-TurboS
CLOSEDTencent · Feb 2025
560B·56Ba256K ctx$0.12/$0.3
1267.4
ReasoningFunction callingTool use
GPT-4.1 mini
CLOSEDOpenAI · Apr 2025
—1M ctx$0.4/$1.6
1256.8
VisionFunction callingTool useAgentic
Gemini 2.5 Flash-Lite (Thinking)
CLOSEDGoogle · Jun 2025
—1M ctx$0.1/$0.4
1256.0
ReasoningVisionFunction callingTool useAgentic
QwQ-32B
OPENAlibaba · Mar 2025
32.5B128K ctxweights
1248.6
ReasoningFunction callingTool use
Claude 3.5 Sonnet
CLOSEDAnthropic · Jun 2024
—200K ctx$3/$15
1247.7
ReasoningVisionFunction callingTool useAgentic
MiniMax-Text-01
OPENMiniMax · Jan 2025
456B·45.9Ba1M ctxweights
1237.0
Function callingTool useAgentic
OpenAI o1-mini
CLOSEDOpenAI · Sep 2024
—128K ctx$3/$12
1227.9
Reasoning
DeepSeek-V3 671B
OPENDeepSeek · Dec 2024
671B·37Ba128K ctx$0.2/$0.8
1227.5
Function callingTool use
MiniMax-VL-01
OPENMiniMax · Jan 2025
456B·45.9Ba1M ctx$0.2/$1.1
1220.3
VisionFunction callingTool use
Reka Flash 3
OPENReka AI · Mar 2025
21B32K ctx$0.2/$0.8
1217.3
ReasoningFunction callingTool useAgentic
Llama 4 Maverick
OPENMeta · Apr 2025
400B·17Ba1M ctxweights
1216.1
VisionFunction callingTool useAgentic
Qwen3-235B-A22B (Non-Thinking)
OPENAlibaba · Apr 2025
235B·22Ba128K ctxweights
1213.0
Function callingTool useAgentic
GPT-4o
CLOSEDOpenAI · May 2024
—128K ctx$2.5/$10
1212.1
VisionFunction callingTool use
LG EXAONE 4.0 32B (Non-reasoning)
OPENLG AI Research · Jul 2025
32B131K ctxweights
1211.0
Function callingTool useAgentic
Qwen2.5-Max
CLOSEDAlibaba · Jan 2025
—32K ctx$1.6/$6.4
1206.3
Function callingTool useAgentic
Tülu 3 405B
OPENAllen Institute for AI (Ai2) · Jan 2025
405.85B128K ctxweights
1201.7
Amazon Nova Premier
CLOSEDAmazon · Apr 2025
—1M ctx$2.5/$12.5
1194.1
VisionFunction callingTool useAgentic
Llama 3.1 405B
OPENMeta · Jul 2024
405B128K ctxweights
1187.9
Function callingTool useAgentic
Mistral Medium 3
CLOSEDMistral AI · May 2025
—128K ctx$0.4/$2
1186.1
VisionFunction callingTool useAgentic
Grok-2
CLOSEDSpaceXAI · Aug 2024
—131K ctx$2/$10
1182.7
VisionFunction callingTool use
Llama 3.3 70B
OPENMeta · Dec 2024
70B128K ctxweights
1182.4
Function callingTool useAgentic
Qwen2.5-VL-72B
OPENAlibaba · Jan 2025
73B32K ctxweights
1171.4
VisionFunction callingTool useAgentic
Gemini 2.5 Flash-Lite (Non-thinking)
CLOSEDGoogle · Jun 2025
—1M ctx$0.1/$0.4
1165.2
ReasoningVisionFunction callingTool useAgentic
Amazon Nova Pro
CLOSEDAmazon · Dec 2024
—300K ctx$0.8/$3.2
1158.2
VisionFunction callingTool useAgentic
Qwen2-Math-72B
OPENAlibaba · Aug 2024
72.7B4K ctxweights
1145.4
Tool use
Command A
OPENCohere · Mar 2025
111B256K ctx$2.5/$10
1125.3
Function callingTool useAgentic
Qwen3-32B (Non-Thinking)
OPENAlibaba · Apr 2025
32.8B128K ctx$0.287/$1.147
1097.1
ReasoningFunction callingTool useAgentic
GPT-4 Turbo
CLOSEDOpenAI · Nov 2023
—128K ctx$10/$30
1096.9
VisionFunction callingTool use
GLM-4-Plus
CLOSEDZhipu AI · Aug 2024
—128K ctx$0.69/$0.69
1093.0
Function callingTool useAgentic
QVQ-72B-Preview
OPENAlibaba · Dec 2024
73B32K ctxweights
1086.6
ReasoningVision
QwQ-32B-Preview
OPENAlibaba · Nov 2024
32.5B32K ctxweights
1084.7
Reasoning
Grok-2 mini
CLOSEDSpaceXAI · Aug 2024
—131K ctx—
1084.6
VisionFunction callingTool use
Llama-3.1-Nemotron-Ultra-253B v1 (Reasoning Off)
OPENNVIDIA · Apr 2025
253B128K ctx$2/$6
1075.6
ReasoningFunction callingTool useAgentic
Llama-3.3-Nemotron-Super-49B v1 (Reasoning Off)
OPENNVIDIA · Mar 2025
49.9B128K ctx$0.1/$0.5
1073.3
ReasoningFunction callingTool useAgentic
Qwen3-30B-A3B (Non-Thinking)
OPENAlibaba · Apr 2025
30.5B·3.3Ba128K ctx$0.108/$0.431
1070.4
ReasoningFunction callingTool useAgentic
Sonar
CLOSEDPerplexity AI · Feb 2025
70B128K ctx$1/$1
1066.1
Function callingTool use
Qwen3-14B (Non-Thinking)
OPENAlibaba · Apr 2025
14.8B128K ctxweights
1060.2
Function callingTool useAgentic
DeepSeek-V2.5
OPENDeepSeek · Sep 2024
236B·21Ba128K ctxweights
1050.8
Function callingTool use
Claude 3 Opus
CLOSEDAnthropic · Mar 2024
—200K ctx$15/$75
1047.3
ReasoningVisionFunction callingTool useAgentic
Gemini 2.0 Flash-Lite
CLOSEDGoogle · Feb 2025
—1M ctx$0.075/$0.3
1041.3
VisionFunction callingTool useAgentic
gpt-oss-20b (Low)
OPENOpenAI · Aug 2025
21B·3.6Ba128K ctxweights
1040.0
ReasoningFunction callingTool useAgentic
Yi-Lightning
CLOSED01.AI · Oct 2024
—16K ctx$0.14/$0.14
1028.4
Mistral Large 2 (24.07)
OPENMistral AI · Jul 2024
123B128K ctxweights
1025.1
Function callingTool useAgentic
Hunyuan-Large (A52B)
OPENTencent · Nov 2024
389B·52Ba128K ctxweights
1020.5
Function callingTool use
Gemini 1.5 Pro
CLOSEDGoogle · Feb 2024
—1M ctx—
1013.3
VisionFunction callingTool useAgentic
Mistral Small 3.2 (24B)
OPENMistral AI · Jun 2025
24B128K ctx$0.08/$0.2
1011.7
VisionFunction callingTool useAgentic
DeepSeek-R1-Distill-Qwen-32B
OPENDeepSeek · Jan 2025
32B128K ctxweights
1003.6
Reasoning
Claude 3.5 Haiku
CLOSEDAnthropic · Oct 2024
—200K ctx$0.8/$4
997.9
VisionFunction callingTool useAgentic
Mistral Small 3.1 (24B)
OPENMistral AI · Mar 2025
24B128K ctx$0.1/$0.3
991.2
VisionFunction callingTool use
Gemini 1.0 Ultra
CLOSEDGoogle · Feb 2024
—32K ctx—
985.9
VisionFunction callingTool use
Llama 4 Scout
OPENMeta · Apr 2025
109B·17Ba10M ctxweights
984.3
VisionFunction callingTool useAgentic
Tülu 3 70B
OPENAllen Institute for AI (Ai2) · Nov 2024
70B128K ctxweights
984.3
Phi-4 (14B)
OPENMicrosoft · Dec 2024
14B16K ctx$0.125/$0.5
984.2
GPT-4
CLOSEDOpenAI · Mar 2023
—8K ctx$30/$60
980.0
VisionFunction callingTool use
DeepSeek-Coder-V2 236B
OPENDeepSeek · Jun 2024
236B·21Ba128K ctx$0.14/$0.28
977.3
Function callingTool use
Amazon Nova Lite
CLOSEDAmazon · Dec 2024
—300K ctx$0.06/$0.24
977.2
VisionFunction callingTool useAgentic
Llama 3.2 90B Vision
OPENMeta · Sep 2024
88.8B128K ctxweights
975.9
VisionFunction callingTool useAgentic
Gemma 3 27B
OPENGoogle · Mar 2025
27.4B128K ctxweights
970.6
VisionFunction callingTool use
Skywork-o1-Open-Llama-3.1-8B
OPENKunlun Tech · Nov 2024
8B128K ctxweights
948.5
Reasoning
Llama 3.1 70B
OPENMeta · Jul 2024
70.6B128K ctxweights
939.1
Function callingTool useAgentic
GPT-4o mini
CLOSEDOpenAI · Jul 2024
—128K ctx$0.15/$0.6
938.8
VisionFunction callingTool useAgentic
Mistral Small 3 (24B)
OPENMistral AI · Jan 2025
24B32K ctx$0.1/$0.3
937.1
Function callingTool use
YandexGPT 5 Pro
CLOSEDYandex · Feb 2025
—32K ctx—
936.2
Function callingTool useAgentic
GLM-4
CLOSEDZhipu AI / THUDM (Tsinghua) · Jan 2024
—128K ctx—
932.8
Function callingTool useAgentic
Qwen2.5-72B
OPENAlibaba · Sep 2024
72.7B128K ctxweights
927.6
ReasoningFunction callingTool useAgentic
Palmyra X5
CLOSEDWriter · Apr 2025
—1M ctx$0.6/$6
926.5
ReasoningFunction callingTool useAgentic
Granite 4.1 8B Instruct
OPENIBM · Apr 2026
8B512K ctxweights
918.6
VisionFunction callingTool useAgentic
Falcon-H1 34B
OPENTII · May 2025
34B256K ctxweights
916.0
Function callingTool useAgentic
Gemma 3 12B
OPENGoogle · Mar 2025
12.2B128K ctxweights
909.3
VisionFunction callingTool use
Qwen3-8B (Non-Thinking)
OPENAlibaba · Apr 2025
8.2B128K ctx$0.072/$0.287
900.6
ReasoningFunction callingTool useAgentic
NAVER HyperCLOVA X SEED Think 32B
OPENNAVER · Dec 2025
32B128K ctxweights
897.4
ReasoningVisionFunction callingTool useAgentic
LG EXAONE 3.5 32B Instruct
OPENLG AI Research · Dec 2024
32B32K ctxweights
872.4
Qwen2.5-Math-72B
OPENAlibaba · Sep 2024
72.7B4K ctxweights
870.7
ReasoningTool use
Granite 4.0 H Small
OPENIBM · Oct 2025
32B·9Ba128K ctx$0.06/$0.25
869.2
Function callingTool useAgentic
Reka Core
CLOSEDReka AI · Apr 2024
—128K ctx$2/$6
867.8
ReasoningVisionFunction callingTool useAgentic
SmolLM3 3B (Think)
OPENHugging Face · Jul 2025
3.08B128K ctxweights
864.2
ReasoningFunction callingTool useAgentic
InternLM3-8B-Instruct
OPENShanghai AI Laboratory · Jan 2025
8B128K ctxweights
858.9
ReasoningFunction callingTool useAgentic
Amazon Nova Micro
CLOSEDAmazon · Dec 2024
—128K ctx$0.035/$0.14
858.0
Function callingTool useAgentic
GPT-4.1 nano
CLOSEDOpenAI · Apr 2025
—1M ctx$0.1/$0.4
851.0
VisionFunction callingTool useAgentic
Nemotron-4 340B
OPENNVIDIA · Jun 2024
340B4K ctxweights
846.4
Nemotron-H 56B
OPENNVIDIA · Apr 2025
56B8K ctxweights
846.0
OLMo 3 7B
OPENAllen Institute for AI (Ai2) · Nov 2025
7.3B64K ctxweights
843.3
Function callingTool useAgentic
Llama 3 70B
OPENMeta · Apr 2024
70B8K ctxweights
835.5
abab6.5
CLOSEDMiniMax · Apr 2024
—200K ctx—
835.2
Function callingTool use
Qwen2-VL-72B
OPENAlibaba · Sep 2024
73.4B32K ctxweights
829.4
Vision
Inflection-2.5
CLOSEDInflection AI · Mar 2024
—8K ctx—
824.7
DeepSeek-V2 236B
OPENDeepSeek · May 2024
236B·21Ba128K ctx$0.14/$0.28
821.7
Function callingTool use
Gemini 1.5 Flash
CLOSEDGoogle · May 2024
—1M ctx$0.075/$0.3
813.9
VisionFunction callingTool use
Qwen2.5-32B
OPENAlibaba · Sep 2024
32.5B128K ctxweights
802.6
ReasoningFunction callingTool useAgentic
Molmo 2 8B
OPENAllen Institute for AI (Ai2) · Dec 2025
8.66B36K ctxweights
797.3
Vision
Qwen2-72B
OPENAlibaba · Jun 2024
72B128K ctxweights
792.3
Function callingTool useAgentic
Hermes 3 405B
OPENNous Research · Aug 2024
405B128K ctx$1/$1
785.9
Function callingTool useAgentic
Claude 3 Sonnet
CLOSEDAnthropic · Mar 2024
—200K ctx$3/$15
777.3
ReasoningVisionFunction callingTool useAgentic
OLMo 2 32B
OPENAllen Institute for AI (Ai2) · Mar 2025
32.23B4K ctxweights
775.7
Claude 2
CLOSEDAnthropic · Jul 2023
—100K ctx$8/$24
767.2
Reasoning
Grok-1.5
CLOSEDSpaceXAI · Mar 2024
—128K ctx—
763.1
Phi-3.5-MoE (16x3.8B, 6.6B active)
OPENMicrosoft · Aug 2024
41.9B·6.6Ba128K ctx$0.13/$0.52
741.4
PaLM 2
CLOSEDGoogle · May 2023
—8K ctx—
738.8
AI21 Jamba 1.5 Large
OPENAI21 Labs · Aug 2024
398B·94Ba256K ctx$2/$8
736.1
Function callingTool useAgentic
Phi-3-medium (14B)
OPENMicrosoft · May 2024
14B128K ctx$0.17/$0.68
714.8
Reasoning
DeepSeekMath 7B
OPENDeepSeek · Feb 2024
7B4K ctxweights
712.5
Reasoning
Upstage Solar Pro
OPENUpstage · Sep 2024
22B4K ctxweights
706.9
Claude 1
CLOSEDAnthropic · Mar 2023
—9K ctx$11.02/$32.68
705.7
Granite 3.3 8B Instruct
OPENIBM · Apr 2025
8.1B128K ctxweights
705.0
ReasoningFunction callingTool use
InternLM2.5-7B
OPENShanghai AI Laboratory · Jul 2024
7B1M ctxweights
688.0
ReasoningFunction callingTool useAgentic
Claude 3 Haiku
CLOSEDAnthropic · Mar 2024
—200K ctx$0.25/$1.25
671.7
ReasoningVisionFunction callingTool useAgentic
SmolLM3 3B (No Thinking)
OPENHugging Face · Jul 2025
3.08B128K ctxweights
671.4
Function callingTool useAgentic
Aya Vision 32B
OPENCohere Labs · Mar 2025
32B16K ctxweights
670.5
Vision
Granite 3.2 8B Instruct
OPENIBM · Feb 2025
8.1B128K ctxweights
662.8
ReasoningFunction callingTool use
Mistral Large
CLOSEDMistral AI · Feb 2024
—32K ctx—
661.8
Function callingTool use
Qwen2.5-14B
OPENAlibaba · Sep 2024
14.7B128K ctxweights
660.4
ReasoningFunction callingTool useAgentic
Mistral Large 24.11
OPENMistral AI · Nov 2024
123B128K ctxweights
653.6
Function callingTool useAgentic
Mixtral 8x22B
OPENMistral AI · Apr 2024
141B·39Ba64K ctxweights
633.8
Function callingTool use
Gemma 3 4B
OPENGoogle · Mar 2025
4.3B128K ctxweights
631.4
VisionFunction callingTool use
Qwen1.5-110B
OPENAlibaba · Apr 2024
111B32K ctxweights
627.5
Function callingTool useAgentic
Llama-3.1-Nemotron-70B Instruct
OPENNVIDIA · Oct 2024
70.6B128K ctx$1.2/$1.2
624.3
Function callingTool use
Llama 3.2 11B Vision
OPENMeta · Sep 2024
10.6B128K ctxweights
620.8
Vision
Phi-3-small (7B)
OPENMicrosoft · May 2024
7B8K ctx$0.15/$0.6
615.7
Inflection-2
CLOSEDInflection AI · Nov 2023
—8K ctx—
597.6
Reka Flash (21B)
CLOSEDReka AI · Feb 2024
21B128K ctx—
593.9
Vision
GLM-4-9B-Chat
OPENZhipu AI · Jun 2024
9B128K ctxweights
591.5
Function callingTool useAgentic
Gemini 1.0 Pro
CLOSEDGoogle · Dec 2023
—32K ctx—
588.7
VisionFunction callingTool use
OLMo 2 13B
OPENAllen Institute for AI (Ai2) · Nov 2024
13B4K ctxweights
575.7
Gemma 2 27B
OPENGoogle · Jun 2024
27.2B8K ctxweights
570.9
Pixtral 12B
OPENMistral AI · Sep 2024
12.4B128K ctxweights
549.2
Vision
Qwen2-57B-A14B
OPENAlibaba · Jun 2024
57B·14Ba64K ctxweights
549.1
Function callingTool useAgentic
DBRX
OPENDatabricks · Mar 2024
132B·36Ba32K ctxweights
535.2
ReasoningFunction callingTool use
Qwen2.5-7B
OPENAlibaba · Sep 2024
7.61B128K ctxweights
533.1
ReasoningFunction callingTool useAgentic
Llama 3.1 8B
OPENMeta · Jul 2024
8.03B128K ctxweights
532.3
Function callingTool useAgentic
MiniCPM3-4B
OPENOpenBMB · Sep 2024
4B32K ctxweights
530.0
Function callingTool useAgentic
Phi-4-mini (3.8B)
OPENMicrosoft · Feb 2025
3.8B128K ctx$0.075/$0.3
511.6
Function callingTool use
AI21 Jamba 1.5 Mini
OPENAI21 Labs · Aug 2024
52B·12Ba256K ctx$0.2/$0.4
508.6
Function callingTool useAgentic
Qwen1.5-72B
OPENAlibaba · Feb 2024
72B32K ctxweights
508.3
Phi-4-mini-instruct
OPENMicrosoft · Feb 2025
3.8B128K ctx$0.075/$0.3
505.6
Function callingTool useAgentic
Aya Expanse 32B
OPENCohere Labs · Oct 2024
32B128K ctx$0.5/$1.5
505.2
Ministral 8B
OPENMistral AI · Oct 2024
8.02B128K ctx$0.1/$0.1
503.5
Function callingTool useAgentic
Qwen-72B
OPENAlibaba · Nov 2023
72B32K ctxweights
501.3
Skywork-MoE-Base
OPENKunlun Tech · Jun 2024
146B·22Ba8K ctxweights
491.9
Yi-1.5 34B
OPEN01.AI · May 2024
34.4B4K ctxweights
491.7
Command R7B
OPENCohere · Dec 2024
7B128K ctx$0.0375/$0.15
491.3
Function callingTool useAgentic
Claude Instant 1.2
CLOSEDAnthropic · Aug 2023
—100K ctx$1.63/$5.51
491.1
Reasoning
Gemini 1.5 Flash-8B
CLOSEDGoogle · Oct 2024
8B1M ctx$0.0375/$0.15
483.5
VisionFunction callingTool use
Grok-1
OPENSpaceXAI · Mar 2024
314B·86Ba8K ctxweights
477.5
Tülu 2+DPO 70B
OPENAllen Institute for AI (Ai2) · Nov 2023
69B8K ctxweights
467.5
DeepSeek LLM 67B
OPENDeepSeek · Nov 2023
67B4K ctxweights
463.4
LG EXAONE 3.0 7.8B Instruct
OPENLG AI Research · Aug 2024
7.8B4K ctxweights
456.3
Phi-3.5-mini (3.8B)
OPENMicrosoft · Aug 2024
3.8B128K ctx$0.13/$0.52
452.2
Qwen2-7B
OPENAlibaba · Jun 2024
7.07B128K ctxweights
448.3
Function callingTool useAgentic
Llama 3 8B
OPENMeta · Apr 2024
8.03B8K ctxweights
443.6
InternLM2-20B
OPENShanghai AI Laboratory · Jan 2024
20B200K ctxweights
443.5
ReasoningFunction callingTool useAgentic
Inflection-1
CLOSEDInflection AI · Jun 2023
—8K ctx—
443.2
Phi-3-mini (3.8B)
OPENMicrosoft · Apr 2024
3.8B4K ctx$0.13/$0.52
440.4
Gemma 2 9B
OPENGoogle · Jun 2024
9.24B8K ctxweights
434.8
Snowflake Arctic
OPENSnowflake · Apr 2024
480B·17Ba4K ctxweights
434.7
NAVER HyperCLOVA X
CLOSEDNAVER · Aug 2023
204B32K ctx—
432.0
Reasoning
Nous Hermes 2 Mixtral 8x7B DPO
OPENNous Research · Jan 2024
46.7B·12.9Ba32K ctxweights
427.7
Function callingTool useAgentic
GPT-3.5 Turbo
CLOSEDOpenAI · Mar 2023
—16K ctx$0.5/$1.5
426.2
Function callingTool use
Krutrim-2 (12B)
OPENOla Krutrim · Jan 2025
12B128K ctxweights
421.7
Falcon3-10B
OPENTII · Dec 2024
10B32K ctxweights
415.4
Function callingTool useAgentic
Codestral 22B
OPENMistral AI · May 2024
22.2B32K ctxweights
415.3
Stable Beluga 2 (FreeWilly 2) 70B
OPENStability AI (CarperAI) · Jul 2023
70B4K ctxweights
412.7
Reasoning
Falcon3-7B
OPENTII · Dec 2024
7B32K ctxweights
406.0
Function callingTool useAgentic
Mixtral 8x7B
OPENMistral AI · Dec 2023
46.7B·12.9Ba32K ctxweights
393.3
Yi-34B
OPEN01.AI · Nov 2023
34B4K ctxweights
389.2
ERNIE Bot (ERNIE 3.5)
CLOSEDBaidu · Jun 2023
—8K ctx—
388.4
Function callingTool use
OLMo 2 7B
OPENAllen Institute for AI (Ai2) · Nov 2024
7.3B4K ctxweights
388.0
Command R
OPENCohere · Mar 2024
35B128K ctx$0.5/$1.5
385.9
Function callingTool useAgentic
Mistral NeMo 12B
OPENMistral AI · Jul 2024
12B128K ctx$0.15/$0.15
375.5
Function callingTool use
Granite 3.0 8B Instruct
OPENIBM · Oct 2024
8.1B4K ctxweights
371.7
Function callingTool use
Ministral 3B
CLOSEDMistral AI · Oct 2024
3B128K ctx$0.04/$0.04
370.2
Function callingTool useAgentic
Airoboros 70B (L2)
OPENJon Durbin (independent) · Aug 2023
70B4K ctxweights
368.5
Function calling
Stable LM 2 12B
OPENStability AI · Apr 2024
12.1B4K ctxweights
359.5
Function callingTool use
StarCoder2 15B
OPENBigCode · Feb 2024
15B16K ctxweights
359.4
Yi-1.5 9B
OPEN01.AI · May 2024
9B4K ctxweights
351.4
Falcon-180B
OPENTII · Sep 2023
180B2K ctxweights
342.4
PaLM (540B)
CLOSEDGoogle · Apr 2022
540B2K ctx—
337.0
Reasoning
Llama 3.2 3B
OPENMeta · Sep 2024
3.21B128K ctxweights
334.6
Function callingTool useAgentic
Reka Edge (7B)
CLOSEDReka AI · Feb 2024
7B64K ctx$0.1/$0.1
332.8
Vision
Starling-7B (alpha)
OPENUC Berkeley (Berkeley NEST) / Nexusflow · Nov 2023
7B8K ctxweights
330.7
Granite 3.1 8B Instruct
OPENIBM · Dec 2024
8.1B128K ctxweights
324.7
Function callingTool use
Upstage SOLAR 10.7B
OPENUpstage · Dec 2023
10.7B4K ctxweights
317.0
ChatGLM3-6B
OPENZhipu AI / THUDM (Tsinghua) · Oct 2023
6.2B8K ctxweights
313.4
Function callingTool useAgentic
Guanaco 65B
OPENUniversity of Washington (Dettmers et al.) · May 2023
65B2K ctxweights
293.0
Llama 2 70B
OPENMeta · Jul 2023
69B4K ctxweights
291.6
Qwen-14B
OPENAlibaba · Sep 2023
14B8K ctxweights
291.4
Llama 2 70B Chat
OPENMeta · Jul 2023
69B4K ctxweights
265.5
Chinchilla
CLOSEDGoogle · Mar 2022
70B2K ctx—
252.0
WizardLM-2 8x22B
OPENMicrosoft · Apr 2024
141B·39Ba64K ctxweights
251.3
Aya 23 35B
OPENCohere Labs · May 2024
35B8K ctxweights
228.8
AI21 Jamba
OPENAI21 Labs · Mar 2024
52B·12Ba256K ctxweights
224.3
Zephyr 7B Beta
OPENHugging Face · Oct 2023
7.24B8K ctxweights
219.5
Orca 2 (13B)
OPENMicrosoft · Nov 2023
13B4K ctxweights
207.7
Reasoning
Nemotron-4 15B
OPENNVIDIA · Feb 2024
15B4K ctxweights
206.7
UL2 20B
OPENGoogle · Oct 2022
20B0.5K ctxweights
204.5
Reasoning
Guanaco 33B
OPENUniversity of Washington (Dettmers et al.) · May 2023
33B2K ctxweights
201.5
CodeGemma 7B
OPENGoogle · Apr 2024
8.54B8K ctxweights
194.5
OpenHermes 2.5 Mistral 7B
OPENNous Research · Nov 2023
7.24B8K ctxweights
186.9
Code Llama 34B
OPENMeta · Aug 2023
34B16K ctxweights
182.8
Falcon 2 11B
OPENTII · May 2024
11B8K ctxweights
182.2
PLaMo-100B
OPENPreferred Networks · Oct 2024
100B4K ctxweights
170.1
DeciLM-7B
OPENDeci AI · Dec 2023
7.04B8K ctxweights
166.3
Phi-2 (2.7B)
OPENMicrosoft · Dec 2023
2.7B2K ctxweights
165.3
StarCoder (15.5B)
OPENBigCode · May 2023
15.5B8K ctxweights
163.2
LLaMA 65B
OPENMeta · Feb 2023
65.2B2K ctxweights
158.4
Falcon Mamba 7B
OPENTII · Aug 2024
7B8K ctxweights
148.0
Granite 3.0 2B Instruct
OPENIBM · Oct 2024
2.53B4K ctxweights
140.3
Function callingTool useAgentic
Falcon-40B
OPENTII · May 2023
40B2K ctxweights
131.5
Gemma 7B
OPENGoogle · Feb 2024
8.54B8K ctxweights
126.7
H2O-Danube3 4B
OPENH2O.ai · Jul 2024
3.96B8K ctxweights
124.0
Command
CLOSEDCohere · Jan 2023
—4K ctx$1/$2
121.1
Mistral 7B
OPENMistral AI · Sep 2023
7.3B8K ctx$0.25/$0.25
115.1
GPT-3 175B
CLOSEDOpenAI · Jun 2020
175B2K ctx—
108.9
MiniCPM-2B
OPENOpenBMB · Feb 2024
2.7B4K ctxweights
107.7
LLaMA 33B
OPENMeta · Feb 2023
32.5B2.048K ctxweights
91.7
Qwen-7B
OPENAlibaba · Aug 2023
7B32K ctxweights
89.3
Skywork-13B
OPENKunlun Tech · Oct 2023
13B4K ctxweights
88.3
Qwen1.5-MoE-A2.7B
OPENAlibaba · Mar 2024
14.3B·2.7Ba32K ctxweights
85.5
StarCoder2 3B
OPENBigCode · Feb 2024
3B16K ctxweights
84.5
Stable Beluga 1 (FreeWilly 1) 65B
OPENStability AI (CarperAI) · Jul 2023
65B2K ctxweights
84.3
Reasoning
Baichuan2-13B
OPENBaichuan Intelligence · Sep 2023
13B4K ctxweights
77.7
Apple DCLM-Baseline 7B
OPENApple · Jul 2024
7B2K ctxweights
62.7
Flan-T5 XXL
OPENGoogle · Oct 2022
11B1K ctxweights
42.3
Reasoning
Vicuna 13B v1.5
OPENLMSYS (UC Berkeley/CMU/Stanford/UCSD) · Aug 2023
13B4K ctxweights
39.9
Llama 2 13B Chat
OPENMeta · Jul 2023
13B4K ctxweights
39.0
Llama-3.1-Minitron 4B
OPENNVIDIA · Aug 2024
4.5B8K ctxweights
36.4
Code Llama 34B Python
OPENMeta · Aug 2023
34B16K ctxweights
34.5
Sarvam-1 (2B)
OPENSarvam AI · Oct 2024
2B8K ctxweights
34.4
Galactica 120B
OPENMeta · Nov 2022
120B2.048K ctxweights
32.8
Gemini 1.0 Nano
CLOSEDGoogle · Dec 2023
3.25B32K ctx—
29.2
Vision
OLMoE-1B-7B-0924
OPENAllen Institute for AI (Ai2) · Sep 2024
6.9B·1.3Ba4K ctxweights
24.1
MPT-30B-Instruct
OPENMosaicML · Jun 2023
30B8K ctxweights
14.9
Llama 3.2 1B
OPENMeta · Sep 2024
1.24B128K ctxweights
14.7
Function callingTool useAgentic
Jais 70B
OPENInception (G42) · Aug 2024
70B4K ctxweights
2.3
Gopher
CLOSEDGoogle · Dec 2021
280B2K ctx—
1.0
WizardLM 7B
OPENMicrosoft & Peking University · Apr 2023
7B2K ctxweights
-10.7
Llama 2 13B
OPENMeta · Jul 2023
13B4K ctxweights
-11.4
MPT-30B
OPENMosaicML · Jun 2023
30B8K ctxweights
-13.6
Yi-6B
OPEN01.AI · Nov 2023
6B4K ctxweights
-15.0
Nemotron-3 8B
OPENNVIDIA · Nov 2023
8B4K ctxweights
-16.8
Vicuna 13B (v1.1)
OPENLMSYS (UC Berkeley/CMU/Stanford/UCSD) · Apr 2023
13B2K ctxweights
-22.8
Poro 34B
OPENSilo AI (LumiOpen) · Apr 2024
34.2B2K ctxweights
-25.6
AI21 Jurassic-2 (Ultra)
CLOSEDAI21 Labs · Mar 2023
—8K ctx$18.8/$18.8
-26.0
DeepSeek-VL 7B
OPENDeepSeek · Mar 2024
7B4K ctxweights
-29.9
Vision
StableLM Zephyr 3B
OPENStability AI · Dec 2023
2.8B4K ctxweights
-42.8
GigaChat
CLOSEDSber (Sberbank) · Apr 2023
—128K ctx—
-42.9
Function callingTool use
Sarashina2-70B
OPENSB Intuitions · Aug 2024
70B8K ctxweights
-56.0
Teuken-7B (instruct)
OPENOpenGPT-X · Nov 2024
7B4K ctxweights
-58.5
Baichuan2-7B
OPENBaichuan Intelligence · Sep 2023
7B4K ctxweights
-66.0
LLaMA 13B
OPENMeta · Feb 2023
13B2.048K ctxweights
-94.4
Llama 2 7B Chat
OPENMeta · Jul 2023
7B4K ctxweights
-104.9
Koala 13B
OPENUC Berkeley (BAIR) · Apr 2023
13B2K ctxweights
-106.9
Comma v0.1-2T
OPENEleutherAI · Jun 2025
7B4K ctxweights
-108.7
DeepSeekMoE 16B
OPENDeepSeek · Jan 2024
16.4B·2.8Ba4K ctxweights
-116.2
SmolLM2 1.7B
OPENHugging Face · Nov 2024
1.71B8K ctxweights
-140.6
GLM-130B
OPENZhipu AI / THUDM (Tsinghua) · Aug 2022
130B2K ctxweights
-144.2
Pharia-1-LLM-7B (control)
OPENAleph Alpha · Aug 2024
7.04B8K ctxweights
-182.4
Phi-1.5 (1.3B)
OPENMicrosoft · Sep 2023
1.3B2K ctxweights
-190.2
BLOOMZ (176B)
OPENBigScience · Nov 2022
176.25B2K ctxweights
-191.8
Llama 2 7B
OPENMeta · Jul 2023
6.7B4K ctxweights
-208.4
Gemma 2B
OPENGoogle · Feb 2024
2.5B8K ctxweights
-208.9
Qwen-1.8B
OPENAlibaba · Nov 2023
1.8B32K ctxweights
-213.0
LaMDA
CLOSEDGoogle · May 2021
137B2K ctx—
-232.9
BLOOM (176B)
OPENBigScience · Jul 2022
176.25B2K ctxweights
-236.7
MPT-7B-Instruct
OPENMosaicML · May 2023
6.7B2.048K ctxweights
-242.5
Stable LM 2 1.6B
OPENStability AI · Jan 2024
1.64B4K ctxweights
-267.5
ChatGLM2-6B
OPENZhipu AI / THUDM (Tsinghua) · Jun 2023
6.24B32K ctxweights
-274.3
Luminous Supreme (70B)
CLOSEDAleph Alpha · Apr 2022
70B2K ctx$175/$192.5
-280.0
OLMo 7B
OPENAllen Institute for AI (Ai2) · Feb 2024
6.9B2K ctxweights
-287.6
LLaMA 7B
OPENMeta · Feb 2023
6.7B2K ctxweights
-288.0
StableLM-3B-4e1t
OPENStability AI · Oct 2023
2.8B4K ctxweights
-288.0
Alpaca 7B
OPENStanford (CRFM / Tatsu Lab) · Mar 2023
6.7B0.512K ctxweights
-301.1
Falcon-7B
OPENTII · May 2023
7B2K ctxweights
-307.6
MPT-7B
OPENMosaicML · May 2023
6.7B2.048K ctxweights
-328.6
H2O-Danube 1.8B
OPENH2O.ai · Jan 2024
1.8B16K ctxweights
-339.0
XGen-7B 8K
OPENSalesforce · Jul 2023
7B8K ctxweights
-371.1
RedPajama-INCITE-7B-Base
OPENTogether · Jun 2023
6.9B2K ctxweights
-404.6
RedPajama-INCITE-7B-Instruct
OPENTogether · Jun 2023
6.9B2K ctxweights
-408.7
OpenLLaMA 7B
OPENOpenLM Research (Berkeley) · May 2023
6.7B2K ctxweights
-421.1
GPT-NeoX-20B
OPENEleutherAI · Feb 2022
20.6B2K ctxweights
-433.1
BTLM-3B-8K
OPENCerebras · Jul 2023
2.6B8K ctxweights
-457.3
Apple OpenELM (270M–3B)
OPENApple · Apr 2024
3B2K ctxweights
-460.4
MPT-7B-StoryWriter-65k+
OPENMosaicML · May 2023
6.7B65K ctxweights
-474.3
GPT-J-6B
OPENEleutherAI · Jun 2021
6.05B2K ctxweights
-479.9
Pythia-6.9B
OPENEleutherAI · Apr 2023
6.9B2K ctxweights
-480.1
RedPajama-INCITE-7B-Chat
OPENTogether · Jun 2023
6.9B2K ctxweights
-483.4
Jais-13B
OPENInception (G42) · Aug 2023
13B2K ctxweights
-488.3
BLOOM-7B1
OPENBigScience · Jul 2022
7.069B2K ctxweights
-490.3
ChatGLM-6B
OPENZhipu AI / THUDM (Tsinghua) · Mar 2023
6.2B2K ctxweights
-505.6
GPT-Neo 2.7B
OPENEleutherAI · Mar 2021
2.7B2K ctxweights
-525.6
Cerebras-GPT 13B
OPENCerebras · Mar 2023
13B2K ctxweights
-546.2
RedPajama-INCITE-Base-3B-v1
OPENTogether · May 2023
2.8B2K ctxweights
-557.8
Pythia-1.4B
OPENEleutherAI · Apr 2023
1.4B2K ctxweights
-565.1
Pythia-12B
OPENEleutherAI · Apr 2023
11.8B2K ctxweights
-566.0
StableLM-Alpha 7B
OPENStability AI · Apr 2023
7B4K ctxweights
-636.7
Claude Mythos 5.1
CLOSEDAnthropic · Sep 2026
—1M ctx—
3523.3
ReasoningVisionFunction callingTool useAgentic
Gemini 3.5 Flash Cyber
CLOSEDGoogle · Jul 2026
—1M ctx—
2940.0
ReasoningFunction callingTool useAgentic
Grok 4.20
CLOSEDSpaceXAI · Mar 2026
—1M ctx$1.25/$2.5
2832.5
ReasoningVisionFunction callingTool useAgentic
DeepSeek-V4-Flash-0731
OPENDeepSeek · Jul 2026
284B·13Ba1M ctx$0.14/$0.28
2697.3
ReasoningFunction callingTool useAgentic
Baichuan 4
CLOSEDBaichuan Intelligence · May 2024
—32K ctx$14.75/$14.75
2538.1
VisionFunction callingTool useAgentic
Laguna S 2.1
OPENPoolside · Jul 2026
118B·8Ba1M ctx$0.1/$0.2
2536.9
ReasoningFunction callingTool useAgentic
ERNIE 4.0
CLOSEDBaidu · Oct 2023
—8K ctx—
2325.9
VisionFunction callingTool use
Claude Opus 4.5 (Medium)
CLOSEDAnthropic · Nov 2025
—200K ctx$5/$25
2255.6
ReasoningVisionFunction callingTool useAgentic
Kimi Chat
CLOSEDMoonshot AI · Oct 2023
—128K ctx$2/$5
2145.6
DeepSeek-Prover-V2 671B
OPENDeepSeek · Apr 2025
671B·37Ba128K ctxweights
2140.6
Reasoning
GPT-5-Codex
CLOSEDOpenAI · Sep 2025
—400K ctx$1.25/$10
2071.8
ReasoningVisionFunction callingTool useAgentic
Step-1V
CLOSEDStepFun · Sep 2024
—32K ctx—
1954.1
VisionFunction callingTool use
Doubao (1st-gen)
CLOSEDByteDance · May 2024
—128K ctx—
1935.8
Function callingTool use
Grok 4.3
CLOSEDSpaceXAI · Apr 2026
—1M ctx$1.25/$2.5
1933.3
ReasoningVisionFunction callingTool useAgentic
Jamba2 Mini
OPENAI21 Labs · Jan 2026
52B·12Ba256K ctxweights
1911.9
Function callingTool use
Laguna XS 2.1
OPENPoolside · Jul 2026
33B·3Ba256K ctx$0.1/$0.2
1818.3
ReasoningFunction callingTool useAgentic
OpenAI o1-pro
CLOSEDOpenAI · Dec 2024
—200K ctx$150/$600
1678.1
ReasoningVisionFunction callingTool useAgentic
Reka Edge (2026)
OPENReka AI · Mar 2026
7B16K ctx$0.1/$0.1
1668.3
VisionTool useAgentic
R1-1776
OPENPerplexity AI · Feb 2025
671B·37Ba128K ctxweights
1544.0
Reasoning
InternVL3-78B
OPENOpenGVLab (Shanghai AI Lab) · Apr 2025
78B32K ctxweights
1478.2
VisionFunction callingTool useAgentic
iFlytek Spark X1 (Xinghuo X1)
CLOSEDiFlytek · Jan 2025
—128K ctx—
1414.1
ReasoningFunction callingTool use
Upstage Solar Pro 2 (Reasoning)
CLOSEDUpstage · Jul 2025
31B64K ctx$0.15/$0.6
1406.2
ReasoningFunction callingTool useAgentic
Devstral Small
OPENMistral AI · May 2025
24B128K ctxweights
1330.2
Function callingTool useAgentic
Palmyra Med 70B
OPENWriter · Jul 2024
70B8K ctxweights
1305.0
Reasoning
ERNIE 4.5 VL 424B-A47B
OPENBaidu · Jun 2025
424B·47Ba128K ctx$0.42/$1.25
1298.5
ReasoningVisionFunction callingTool useAgentic
Gorilla 7B
OPENUC Berkeley & Microsoft Research · May 2023
7B2K ctxweights
1280.7
Function callingTool use
Llama-3.1-ELYZA-JP-70B
CLOSEDELYZA · Oct 2024
70B128K ctx—
1148.5
Function callingTool use
Pixtral Large (124B)
OPENMistral AI · Nov 2024
124B128K ctxweights
1123.9
VisionFunction callingTool useAgentic
Upstage Solar Pro 2 (Chat)
CLOSEDUpstage · Jul 2025
31B64K ctx$0.15/$0.6
1092.8
ReasoningFunction callingTool useAgentic
Molmo 72B
OPENAllen Institute for AI (Ai2) · Sep 2024
72B4K ctxweights
1032.1
Vision
Codestral 25.01
CLOSEDMistral AI · Jan 2025
—256K ctx$0.3/$0.9
1026.6
Amazon Titan Text (Express & Lite)
CLOSEDAmazon · Nov 2023
—8K ctx$0.8/$1.6
—
Apple Foundation Model 3 Core Advanced (on-device, 20B sparse)
CLOSEDApple · Jun 2026
20B·4Ba32K ctx—
—
ReasoningVisionFunction callingTool useAgentic
Apple Foundation Model 3 Core (on-device, ~3B)
CLOSEDApple · Jun 2026
3B32K ctx—
—
VisionFunction callingTool useAgentic
Dolphin 2.5 Mixtral 8x7B
OPENCognitive Computations (Eric Hartford) · Dec 2023
46.7B·12.9Ba16K ctxweights
—
ERNIE 3.0 Titan
CLOSEDBaidu · Dec 2021
260B0.512K ctx—
—
ERNIE X1
CLOSEDBaidu · Mar 2025
—8K ctx$0.28/$1.1
—
ReasoningVisionFunction callingTool useAgentic
Gemini 3.8 Flash Cyber
CLOSEDGoogle · Sep 2026
—1M ctx—
1000.0
ReasoningVisionFunction callingTool useAgentic
GLM-5.2 (High)
OPENZhipu AI / Z.ai (Tsinghua) · Jun 2026
744B·40Ba1M ctx$1.4/$4.4
—
ReasoningFunction callingTool useAgentic
GPT-5.6 Cyber
CLOSEDOpenAI · Aug 2026
—400K ctx$12.5/$75
—
ReasoningVisionFunction callingTool useAgentic
Granite 4.2 8B
OPENIBM · Aug 2026
8B128K ctxweights
—
ReasoningFunction callingTool useAgentic
Grok 4.20 Multi-Agent
CLOSEDSpaceXAI · Mar 2026
—1M ctx$1.25/$2.5
—
ReasoningVisionFunction callingTool useAgentic
Hunyuan (1st-gen)
CLOSEDTencent · Sep 2023
—4K ctx—
—
iFlytek Spark V1.0 (Xinghuo)
CLOSEDiFlytek · May 2023
—8K ctx—
—
iFlytek Spark V4.0 (Xinghuo)
CLOSEDiFlytek · Jun 2024
—8K ctx—
—
VisionFunction callingTool useAgentic
Kimi K3 Fast
CLOSEDMoonshot AI · Jul 2026
2800B1.024M ctx$4.5/$22.5
—
ReasoningVisionFunction callingTool useAgentic
Ling 3.0 Tiny
CLOSEDinclusionAI · Aug 2026
7.9B·1.3Ba256K ctx—
—
ReasoningFunction callingTool use
MAI-1-preview
CLOSEDMicrosoft · Aug 2025
—128K ctx—
—
Qwen3.7 Flash
CLOSEDAlibaba · Jul 2026
—1M ctx$0.03/$0.13
—
ReasoningVisionFunction callingTool useAgentic
rinna japanese-gpt-neox-3.6b
OPENrinna · May 2023
3.6B2K ctxweights
—
Step-1
CLOSEDStepFun · Mar 2024
100B256K ctx—
—
Step-2
CLOSEDStepFun · Jul 2024
1000B16K ctx—
—
iFlytek Spark V3.0 (Xinghuo)
CLOSEDiFlytek · Oct 2023
—8K ctx—
995.0
VisionFunction callingTool useAgentic
MiniCPM-o 2.6
OPENOpenBMB · Jan 2025
8B32K ctxweights
988.9
Vision
Baichuan 3
CLOSEDBaichuan Intelligence · Jan 2024
—32K ctx$1.77/$1.77
973.4
Palmyra X 004
CLOSEDWriter · Oct 2024
—128K ctx$2.5/$10
947.0
ReasoningFunction callingTool useAgentic
DeepSeek-VL2
OPENDeepSeek · Dec 2024
27B·4.5Ba4K ctxweights
878.5
Vision
Qwen-Max-0428
CLOSEDAlibaba · May 2024
—32K ctx—
866.1
Function callingTool useAgentic
Mistral Medium (original)
CLOSEDMistral AI · Dec 2023
—32K ctx$2.75/$8.1
796.2
DeepSeek Coder 33B
OPENDeepSeek · Nov 2023
33.3B16K ctxweights
794.0
Phi-4-multimodal (5.6B)
OPENMicrosoft · Feb 2025
5.6B128K ctx$0.08/$0.32
782.5
VisionFunction callingTool use
Phi-4-multimodal-instruct
OPENMicrosoft · Feb 2025
5.6B128K ctx$0.08/$0.32
782.5
VisionFunction callingTool useAgentic
Qwen2.5-Coder-32B
OPENAlibaba · Nov 2024
32.5B128K ctxweights
743.4
ReasoningFunction callingTool useAgentic
Codestral Mamba (7B)
OPENMistral AI · Jul 2024
7.3B256K ctxweights
734.3
Soofi S
OPENSOOFI Consortium · Jul 2026
31.6B·3.5Ba1.024M ctxweights
719.5
WizardCoder-Python-34B V1.0
OPENMicrosoft · Aug 2023
34B16K ctxweights
709.1
MiniCPM-V 2.6
OPENOpenBMB · Aug 2024
8B32K ctxweights
707.7
Vision
GLM-4V-9B
OPENZhipu AI · Jun 2024
14B8K ctxweights
590.7
Vision
MiniCPM-Llama3-V 2.5
OPENOpenBMB · May 2024
8B8K ctxweights
575.7
Vision
Code Llama 70B Instruct
OPENMeta · Jan 2024
70B16K ctxweights
569.8
Claude 2.1
CLOSEDAnthropic · Nov 2023
—200K ctx$8/$24
532.0
ReasoningFunction callingTool use
Idefics2 8B
OPENHugging Face · Apr 2024
8B32K ctxweights
508.8
Vision
CodeQwen1.5-7B
OPENAlibaba · Apr 2024
7B64K ctxweights
478.0
Phi-3.5-vision (4.2B)
OPENMicrosoft · Aug 2024
4.2B128K ctx$0.13/$0.52
467.1
Vision
CogVLM-17B
OPENZhipu AI / THUDM (Tsinghua) · Oct 2023
17B4K ctxweights
465.4
Vision
Granite 34B Code Instruct
OPENIBM · May 2024
34B8K ctxweights
424.5
Function callingTool useAgentic
Code Llama 70B
OPENMeta · Jan 2024
70B16K ctxweights
418.1
Phi-1 (1.3B)
OPENMicrosoft · Jun 2023
1.3B2K ctxweights
328.3
Qwen-VL
OPENAlibaba · Aug 2023
9.6B8K ctxweights
298.5
Vision
Granite 8B Code Instruct
OPENIBM · May 2024
8B4K ctxweights
296.4
Function callingTool useAgentic
Llama-3-ELYZA-JP-8B
OPENELYZA · Jun 2024
8B8K ctxweights
268.9
Code Llama 13B
OPENMeta · Aug 2023
13B16K ctxweights
182.4
Apple Foundation Model (on-device, ~3B)
CLOSEDApple · Jun 2024
3B32K ctx—
176.5
Tool useAgentic
Megatron-Turing NLG 530B
CLOSEDNVIDIA · Oct 2021
530B2K ctx—
139.6
CodeGen2.5 7B
OPENSalesforce · Jul 2023
7B2K ctxweights
97.4
Code Llama 7B
OPENMeta · Aug 2023
7B16K ctxweights
92.1
Viking 7B
OPENSilo AI (LumiOpen) · May 2024
7.55B4K ctxweights
68.9
CodeT5+ 16B
OPENSalesforce · May 2023
16B2K ctxweights
41.1
CodeGen 16B (Mono)
OPENSalesforce · Mar 2022
16.1B2K ctxweights
-26.4
OpenAI Codex
CLOSEDOpenAI · Aug 2021
12B4K ctx—
-45.0
SantaCoder 1.1B
OPENBigCode · Dec 2022
1.1B2K ctxweights
-153.5
CodeGeeX (13B)
OPENZhipu AI / THUDM (Tsinghua) · Sep 2022
13B2K ctxweights
-207.3
OPT-175B
OPENMeta · May 2022
175B2.048K ctxweights
-222.2
IDEFICS 80B
OPENHugging Face · Aug 2023
80B2K ctxweights
-244.6
Vision
OPT-IML 30B
OPENMeta · Dec 2022
30B2.048K ctxweights
-273.5
OPT-66B
OPENMeta · Jun 2022
66B2.048K ctxweights
-347.5
Aya 101
OPENCohere Labs · Feb 2024
13B1K ctxweights
-349.2
Turing-NLG (17B)
CLOSEDMicrosoft · Feb 2020
17B1K ctx—
-540.1