Noometry

Model comparison

Claude Sonnet 4 vs Llama-3.3-70B-Instruct

Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 39× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.

Last verified . 31 shared benchmarks.

Claude Sonnet 4 Anthropic

40.8

Rank #145 Confirmed

Llama-3.3-70B-Instruct Meta

30.6

Rank #291 Confirmed

Summary

  • They share 31 benchmarks with published results for both. Claude Sonnet 4 scores higher in 9 categories and Llama-3.3-70B-Instruct in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Sonnet 4 leads 43.3 to 15.3.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 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 $3 / $15 for Claude Sonnet 4.
  • Claude Sonnet 4 accepts more context: 200K tokens versus 128K.
  • Llama-3.3-70B-Instruct has downloadable open weights; the other is API-only.

Side by side

Claude Sonnet 4Llama-3.3-70B-Instruct
ProviderAnthropicMeta
Noometry Index40.830.6
Released2025-05-222024-12-06
WeightsProprietaryOpen
Context window200K128K
Max output64K4K
Input $ / M tokens$3$0.10
Output $ / M tokens$15$0.32
Results tracked5843

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

Category by category

Coding Claude Sonnet 4 leads

Claude Sonnet 4: 43.5 (#88), Llama-3.3-70B-Instruct: 31.0 (#290)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
SciCode40%26%
WeirdML46.1%14.4%
LMArena Coding14141268
SWE-bench Verified (bash only)64.9%—
Aider Polyglot61.3%—
GSO4.9%—
BigCodeBench Instruct—46.9%
LiveBench Coding—36.6%
BigCodeBench Complete—57.5%
ALE-Bench655.35—

Agentic & Tool Use Claude Sonnet 4 leads

Claude Sonnet 4: 38.5 (#31), Llama-3.3-70B-Instruct: 25.8 (#105)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
Berkeley Function Calling Leaderboard—31.9%
TheAgentCompany33.1%—
Cybench35%—
DeepResearch Bench46.6%—
OSWorld43.9%—
BALROG—23%
METR Time Horizons62%—

Reasoning Claude Sonnet 4 leads

Claude Sonnet 4: 22.9 (#187), Llama-3.3-70B-Instruct: 14.1 (#327)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
SimpleBench45.5%19.9%
CritPt0.3%0%
LMArena Hard Prompts13721257
DTBench77.1%59.5%
LMCA29%17.5%
Epoch Capabilities Index141.69127.33
ForecastBench60.258.6
ARC-AGI-25.9%—
Kagi LLM Benchmark73%—
ARC-AGI-140%—
EnigmaEval3.1%—
LiveBench Reasoning—50.8%
LiveBench Data Analysis—49.5%
LiveBench—50.2%

Math Claude Sonnet 4 leads

Claude Sonnet 4: 43.3 (#80), Llama-3.3-70B-Instruct: 15.3 (#298)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
OTIS Mock AIME 2024-202571.1%5.1%
LMArena Math13751267
MATH Level 584.4%41.6%
Omni-MATH60.2%—
LiveBench Math—42.2%
FrontierMath (Feb 2025 set)4.1%—
FrontierMath Tier 4 (v1)0%—

Knowledge Claude Sonnet 4 leads

Claude Sonnet 4: 41.8 (#108), Llama-3.3-70B-Instruct: 30.6 (#226)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
GPQA Diamond79.2%47.4%
Confabulations13.2%22.8%
Vectara Hallucination Rate10.3%4.1%
LMArena Expert13721225
Humanity's Last Exam7.8%—
MMLU-Pro84.3%—
GPQA (HELM)70.6%—
MMLU—86.3%

Multimodal Not comparable

Claude Sonnet 4: 26.2 (#121), Llama-3.3-70B-Instruct: —

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
LMArena Vision1191—
GeoBench37%—
VPCT34%—
MindCube44.8%—

Multilingual Claude Sonnet 4 leads

Claude Sonnet 4: 46.7 (#156), Llama-3.3-70B-Instruct: 39.9 (#220)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
LMArena Non-English13331236
LMArena Chinese13501217
LMArena French13631281
LMArena German13311251
LMArena Japanese13021150
LMArena Korean12911143
LMArena Russian13551252
LMArena Spanish13571270

Instruction Following Too close to call

Claude Sonnet 4: 71.7 (#145), Llama-3.3-70B-Instruct: 71.1 (#157)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
LMArena Instruction Following13761242
LiveBench Instruction Following—82.7%
IFEval84%—

Long Context Claude Sonnet 4 leads

Claude Sonnet 4: 33.7 (#259), Llama-3.3-70B-Instruct: 26.4 (#295)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
Fiction.LiveBench46.9%33.3%
LMArena Longer Query13981256

Writing & Preference Claude Sonnet 4 leads

Claude Sonnet 4: 57.1 (#132), Llama-3.3-70B-Instruct: 47.6 (#207)

BenchmarkClaude Sonnet 4Llama-3.3-70B-Instruct
LMArena Text13511274
LMArena Creative Writing13451250
LMArena Multi-Turn13761280
Short-Story Creative Writing81.4%—
EQ-Bench Creative Writing1483—
WildBench83.8%—
LiveBench Language—39.2%

Frequently asked questions

Is Claude Sonnet 4 better than Llama-3.3-70B-Instruct?

Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 30.6 on the Noometry Index. Llama-3.3-70B-Instruct costs 39× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.

Which is cheaper, Claude Sonnet 4 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; Claude Sonnet 4 lists at $3 and $15.

Is Claude Sonnet 4 or Llama-3.3-70B-Instruct better for coding?

Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 31.0 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 4 does, with 200K tokens against 128K.

How many benchmarks do Claude Sonnet 4 and Llama-3.3-70B-Instruct share?

31 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Llama-3.3-70B-Instruct has 43.

Related comparisons

Go deeper