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

Reka Flash 3

  • Reka AI
  • Release · March 10, 2025
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#243 of 266
  1. 5.85.8Phi-4 Multimodal Instruct
  2. 5.75.7MiniCPM-V 4.6 1.3B
  3. 5.75.7Jamba Reasoning 3B
  4. 5.65.6Olmo 3 7B Think
  5. 5.65.6Molmo 7B-D
  6. 5.65.6Reka Flash 3
  7. 5.55.5Ministral 3 8B
  8. 5.45.4Qwen3.5 0.8B (Non-reasoning)
  9. 5.45.4Llama 3.2 Instruct 11B (Vision)
  10. 5.35.3Exaone 4.0 1.2B (Reasoning)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy13.7%#192
  • Humanity's Last Exam4.4%#234
  • AA-LCR0%#229
  • CritPt0%#143
  • GPQA Diamond52.9%#163
  • IFBench30.4%#125
  • τ²-Bench Telecom0%#143
  • Terminal-Bench Hard0%#121
Reka Flash 3Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #101 of 162
  1. 9393Grok 4.7 (Xhigh)
  2. 9292GPT-6 Luna (Medium)
  3. 9191GLM-5.3 (Low)
  4. 9191Gemma 4 26B A4B (Reasoning)
  5. 8989Llama 3.3 Instruct 70B
  6. 8888Reka Flash 3
  7. 8686Grok 4.7 (High)
  8. 8383Claude Opus 5.5 (Medium)
  9. 8282Claude Opus 5.5 (Xhigh)
  10. 8282Claude Opus 5.5 (Low)
Data from Artificial Analysis

Latency

seconds to first token · #58 of 162
  1. 1.31.3Llama 3.2 Instruct 11B (Vision)
  2. 1.41.4MiniMax-M3.1-Flash
  3. 1.41.4Claude Sonnet 5.5 (Medium)
  4. 1.41.4GPT-5.6 Terra (Low)
  5. 1.41.4Agnes 2.5 Pro Beta
  6. 1.41.4Reka Flash 3
  7. 1.51.5GPT-5.6 Terra (Medium)
  8. 1.51.5Magistral Small 1.2
  9. 1.61.6DeepSeek V4 Flash Vision (Max)
  10. 1.61.6Inkling Small
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #67 of 169
  1. $0.30.3NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
  2. $0.310.31Nemotron 3.5 Super VL
  3. $0.330.33Celeris-1
  4. $0.350.35Llama 3.2 Instruct 11B (Vision)
  5. $0.350.35Qwen3 Next 80B A3B Instruct
  6. $0.350.35Reka Flash 3
  7. $0.380.38Mercury 2.5
  8. $0.390.39Trinity Large Thinking
  9. $0.420.42Llama 4 Maverick
  10. $0.430.43Qwen3 Omni 30B A3B (Reasoning)
Data from Artificial Analysis

Nearby models

  • Phi-4 Multimodal InstructMicrosoft5.8
  • MiniCPM-V 4.6 1.3BOpenBMB5.7
  • Jamba Reasoning 3BAI21 Labs5.7
  • Olmo 3 7B ThinkAllen Institute for AI5.6
  • Molmo 7B-DAllen Institute for AI5.6
  • Ministral 3 8BMistral5.5
  • Qwen3.5 0.8B (Non-reasoning)Alibaba5.4
  • Llama 3.2 Instruct 11B (Vision)Meta5.4
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend