Noometry

Model comparison

GPT-5 Mini vs Llama-3.3-70B-Instruct

GPT-5 Mini is the stronger model overall, scoring 41.8 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.4× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.

Last verified . 31 shared benchmarks.

GPT-5 Mini OpenAI

41.8

Rank #128 Confirmed

Llama-3.3-70B-Instruct Meta

30.6

Rank #291 Confirmed

Summary

  • They share 31 benchmarks with published results for both. GPT-5 Mini scores higher in 9 categories and Llama-3.3-70B-Instruct in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where GPT-5 Mini leads 46.7 to 15.3.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 5.1% for Llama-3.3-70B-Instruct.
  • Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
  • GPT-5 Mini accepts more context: 400K tokens versus 128K.
  • Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.

Side by side

GPT-5 MiniLlama-3.3-70B-Instruct
ProviderOpenAIMeta
Noometry Index41.830.6
Released2025-08-072024-12-06
WeightsProprietaryOpen
Context window400K128K
Max output128K4K
Input $ / M tokens$0.25$0.10
Output $ / M tokens$2$0.32
Results tracked6043

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-5 Mini leads

GPT-5 Mini: 40.1 (#146), Llama-3.3-70B-Instruct: 31.0 (#290)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
SciCode39.2%26%
WeirdML52.7%14.4%
LMArena Coding14061268
SWE-bench Verified64.7%—
SWE-bench Verified (bash only)59.8%—
SWE-bench Multilingual39.7%—
BigCodeBench Instruct—46.9%
LiveBench Coding—36.6%
BigCodeBench Complete—57.5%
ALE-Bench799.77—
AlgoTune1.38—

Agentic & Tool Use GPT-5 Mini leads

GPT-5 Mini: 31.1 (#70), Llama-3.3-70B-Instruct: 25.8 (#105)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
Berkeley Function Calling Leaderboard55.5%31.9%
Terminal-Bench34.8%—
BALROG—23%
Vending-Bench 2-31.18—

Reasoning GPT-5 Mini leads

GPT-5 Mini: 23.9 (#168), Llama-3.3-70B-Instruct: 14.1 (#327)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
CritPt0%0%
LMArena Hard Prompts13801257
DTBench80.5%59.5%
LMCA34.2%17.5%
Epoch Capabilities Index145.52127.33
ForecastBench6158.6
ARC-AGI-24.4%—
SimpleBench—19.9%
Kagi LLM Benchmark70.3%—
ARC-AGI-154.3%—
Chess Puzzles30%—
EnigmaEval8.2%—
LiveBench Reasoning—50.8%
Mystery Game Puzzles10%—
LiveBench Data Analysis—49.5%
LiveBench—50.2%

Math GPT-5 Mini leads

GPT-5 Mini: 46.7 (#69), Llama-3.3-70B-Instruct: 15.3 (#298)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
OTIS Mock AIME 2024-202586.7%5.1%
LMArena Math13781267
MATH Level 597.8%41.6%
FrontierMath (Tiers 1-3)46.7%—
FrontierMath Tier 412.2%—
ProofBench9%—
Omni-MATH72.2%—
LiveBench Math—42.2%
FrontierMath (Feb 2025 set)27.2%—
FrontierMath Tier 4 (v1)6.3%—

Knowledge GPT-5 Mini leads

GPT-5 Mini: 45.6 (#86), Llama-3.3-70B-Instruct: 30.6 (#226)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
GPQA Diamond75%47.4%
Confabulations13.3%22.8%
Vectara Hallucination Rate12.9%4.1%
LMArena Expert13791225
Humanity's Last Exam19.4%—
SimpleQA Verified21.6%—
MMLU-Pro83.5%—
GPQA (HELM)75.6%—
MMLU—86.3%

Multimodal Not comparable

GPT-5 Mini: 35.6 (#85), Llama-3.3-70B-Instruct: —

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
LMArena Vision1202—
VPCT40.2%—

Multilingual GPT-5 Mini leads

GPT-5 Mini: 48.9 (#137), Llama-3.3-70B-Instruct: 39.9 (#220)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
LMArena Non-English13631236
LMArena Chinese13851217
LMArena French13861281
LMArena German13661251
LMArena Japanese13411150
LMArena Korean13081143
LMArena Russian13621252
LMArena Spanish13551270

Instruction Following GPT-5 Mini leads

GPT-5 Mini: 76.2 (#46), Llama-3.3-70B-Instruct: 71.1 (#157)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
LMArena Instruction Following13571242
LiveBench Instruction Following—82.7%
IFEval92.7%—

Long Context GPT-5 Mini leads

GPT-5 Mini: 41.9 (#132), Llama-3.3-70B-Instruct: 26.4 (#295)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
Fiction.LiveBench69.4%33.3%
LMArena Longer Query13551256

Writing & Preference GPT-5 Mini leads

GPT-5 Mini: 55.2 (#148), Llama-3.3-70B-Instruct: 47.6 (#207)

BenchmarkGPT-5 MiniLlama-3.3-70B-Instruct
LMArena Text13731274
LMArena Creative Writing13251250
LMArena Multi-Turn13631280
Short-Story Creative Writing83.1%—
EQ-Bench Creative Writing1313—
WildBench85.5%—
LiveBench Language—39.2%

Frequently asked questions

Is GPT-5 Mini better than Llama-3.3-70B-Instruct?

GPT-5 Mini is the stronger model overall, scoring 41.8 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 4.4× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.

Which is cheaper, GPT-5 Mini or Llama-3.3-70B-Instruct?

Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; GPT-5 Mini lists at $0.25 and $2.

Is GPT-5 Mini or Llama-3.3-70B-Instruct better for coding?

GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 31.0 in the Noometry coding category.

Which has the bigger context window?

GPT-5 Mini does, with 400K tokens against 128K.

How many benchmarks do GPT-5 Mini and Llama-3.3-70B-Instruct share?

31 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Llama-3.3-70B-Instruct has 43.

Related comparisons

Go deeper