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

Llama 3.3 Instruct 70B

  • Meta
  • Release · December 6, 2024
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#199 of 266
  1. 7.97.9Hermes 4 - Llama-3.1 70B (Reasoning)
  2. 7.87.8Falcon-H1R-7B
  3. 7.87.8Qwen3 Omni 30B A3B (Reasoning)
  4. 7.87.8Gemma 4 E2B (Reasoning)
  5. 7.77.7Step3 VL 10B
  6. 7.77.7Llama 3.3 Instruct 70B
  7. 7.57.5Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
  8. 7.57.5ERNIE 4.5 300B A47B
  9. 7.57.5NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
  10. 7.57.5Gemma 4 E4B (Non-reasoning)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy19%#122
  • Humanity's Last Exam3.6%#256
  • AA-LCR15.7%#204
  • CritPt0%#143
  • GPQA Diamond49.8%#170
  • IFBench47.1%#67
  • τ²-Bench Telecom26.6%#87
  • Terminal-Bench Hard3%#98
Llama 3.3 Instruct 70BMedian
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #100 of 162
  1. 9494Qwen3 Coder Next
  2. 9393Grok 4.7 (Xhigh)
  3. 9292GPT-6 Luna (Medium)
  4. 9191GLM-5.3 (Low)
  5. 9191Gemma 4 26B A4B (Reasoning)
  6. 8989Llama 3.3 Instruct 70B
  7. 8888Reka Flash 3
  8. 8686Grok 4.7 (High)
  9. 8383Claude Opus 5.5 (Medium)
  10. 8282Claude Opus 5.5 (Xhigh)
Data from Artificial Analysis

Latency

seconds to first token · #66 of 162
  1. 1.61.6DeepSeek V4 Flash Vision (Max)
  2. 1.61.6Inkling Small
  3. 1.61.6Sarvam 105B (High)
  4. 1.71.7Claude Fable 5.1 (Low)
  5. 1.81.8DeepSeek V4 Pro 0813 (Max)
  6. 1.81.8Llama 3.3 Instruct 70B
  7. 1.81.8DeepSeek V4 Pro 0813 (Non-reasoning)
  8. 1.81.8Mistral Large 4 Preview
  9. 1.81.8Inkling (Xhigh)
  10. 1.81.8Qwen3.5 Omni Flash
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #85 of 169
  1. $0.530.53Muse Glimmer (High)
  2. $0.540.54MiMo-V2.6-Pro
  3. $0.60.6Beam 1.0
  4. $0.640.64Nemotron 3 Nano Omni 30B A3B Reasoning
  5. $0.660.66DeepSeek V4 Flash Vision (Max)
  6. $0.670.67Llama 3.3 Instruct 70B
  7. $0.70.7Qwen3.7 Plus
  8. $0.750.75Magistral Small 1.2
  9. $0.840.84Qwen3.6 35B A3B (Reasoning)
  10. $0.840.84Qwen3.6 35B A3B (Non-reasoning)
Data from Artificial Analysis

Nearby models

  • Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research7.9
  • Falcon-H1R-7BTII UAE7.8
  • Qwen3 Omni 30B A3B (Reasoning)Alibaba7.8
  • Gemma 4 E2B (Reasoning)Google7.8
  • Step3 VL 10BStepFun7.7
  • Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA7.5
  • ERNIE 4.5 300B A47BBaidu7.5
  • NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA7.5
Data from Artificial Analysis

Same lab

  • Muse Wonder Full (Max)Meta55.8
  • Muse Wonder (Max)Meta52.4
  • Muse Spark Open (Xhigh)Meta38.9
  • Muse Glimmer (High)Meta17.5
  • Llama 4 MaverickMeta9.3
  • Llama 3.1 Instruct 405BMeta7.3
  • Llama 4 ScoutMeta6.5
  • Llama 3.2 Instruct 90B (Vision)Meta6.4
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend