Vascend
HorizonVascend IDEVascend CLICommunityBenchmarkContact
Sign inJoin Discord
HomeHorizonIDECLISign in
ChangelogIntelligenceCost per taskCoding Agent IndexCyber IndexDecision IndexBenchmarksSpeedLatencyPriceLeaderboard
Benchmark

Sarvam 105B (High)

  • Sarvam
  • Release · March 6, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#186 of 266
  1. 99K2-V2 (Medium)
  2. 99Tri-21B-Think
  3. 8.98.9Gemma 4 E4B (Reasoning)
  4. 8.98.9NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
  5. 8.88.8MiniCPM5-1B (Reasoning)
  6. 8.88.8Sarvam 105B (High)
  7. 8.78.7MiniCPM5-1B (Non-reasoning)
  8. 8.78.7Nova 2.0 Lite (Non-reasoning)
  9. 8.68.6Magistral Small 1.2
  10. 8.48.4Nanbeige4.1-3B
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy17.6%#144
  • Humanity's Last Exam11%#141
  • AA-LCR0%#229
  • CritPt0%#143
  • GPQA Diamond73.8%#101
  • IFBench34.4%#109
  • τ²-Bench Telecom46.8%#67
  • Terminal-Bench Hard1.5%#108
Sarvam 105B (High)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #23 of 162
  1. 223223Inkling Small
  2. 213213Ling 3.1 Flash
  3. 209209Ministral 3 3B
  4. 202202NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
  5. 195195Inkling (Xhigh)
  6. 189189Sarvam 105B (High)
  7. 186186Nova 2.0 Lite (High)
  8. 186186gpt-oss-20b (Low)
  9. 186186Claude Haiku 5.5 (Xhigh)
  10. 183183Nova 2.0 Lite (Non-reasoning)
Data from Artificial Analysis

Latency

seconds to first token · #63 of 162
  1. 1.41.4Reka Flash 3
  2. 1.51.5GPT-5.6 Terra (Medium)
  3. 1.51.5Magistral Small 1.2
  4. 1.61.6DeepSeek V4 Flash Vision (Max)
  5. 1.61.6Inkling Small
  6. 1.61.6Sarvam 105B (High)
  7. 1.71.7Claude Fable 5.1 (Low)
  8. 1.81.8DeepSeek V4 Pro 0813 (Max)
  9. 1.81.8Llama 3.3 Instruct 70B
  10. 1.81.8DeepSeek V4 Pro 0813 (Non-reasoning)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #14 of 169
  1. $0.050.05Granite 4.2 3B
  2. $0.060.06Qwen3.5 4B (Reasoning)
  3. $0.060.06Qwen3.5 4B (Non-reasoning)
  4. $0.060.06Nova Micro
  5. $0.070.07NVIDIA Nemotron Nano 9B V2 (Reasoning)
  6. $0.070.07Sarvam 105B (High)
  7. $0.090.09NVIDIA Nemotron Nano 9B V2 (Non-reasoning)
  8. $0.090.09NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
  9. $0.090.09NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)
  10. $0.090.09gpt-oss-20b (High)
Data from Artificial Analysis

Nearby models

  • K2-V2 (Medium)Institute of Foundation Models9
  • Tri-21B-ThinkTrillion Labs9
  • Gemma 4 E4B (Reasoning)Google8.9
  • NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA8.9
  • MiniCPM5-1B (Reasoning)OpenBMB8.8
  • MiniCPM5-1B (Non-reasoning)OpenBMB8.7
  • Nova 2.0 Lite (Non-reasoning)Amazon8.7
  • Magistral Small 1.2Mistral8.6
Data from Artificial Analysis

Same lab

  • Sarvam 30B (High)Sarvam6.6
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend