Noometry

Model comparison

Claude Fable 5 vs Llama 3.1-8B

Claude Fable 5 is the stronger model overall, scoring 66.8 to 23.0 on the Noometry Index. Llama 3.1-8B costs 348× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Last verified . 27 shared benchmarks.

Claude Fable 5 Anthropic

66.8

Rank #5 Confirmed

Llama 3.1-8B Meta

23.0

Rank #352 Confirmed

Summary

  • They share 27 benchmarks with published results for both. Claude Fable 5 scores higher in 9 categories and Llama 3.1-8B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Fable 5 leads 88.5 to 10.2.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5 and 1.7% for Llama 3.1-8B.
  • Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $10 / $50 for Claude Fable 5.
  • Claude Fable 5 accepts more context: 1M tokens versus 128K.
  • Llama 3.1-8B has downloadable open weights; the other is API-only.

Side by side

Claude Fable 5Llama 3.1-8B
ProviderAnthropicMeta
Noometry Index66.823.0
Released2026-06-072024-07-23
WeightsProprietaryOpen
Context window1M128K
Max output128K4K
Input $ / M tokens$10$0.05
Output $ / M tokens$50$0.08
Results tracked6243

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

Category by category

Coding Claude Fable 5 leads

Claude Fable 5: 70.6 (#4), Llama 3.1-8B: 20.2 (#340)

BenchmarkClaude Fable 5Llama 3.1-8B
SciCode61%13.2%
WeirdML91.9%1.7%
LMArena Coding15191195
DeepSWE69.9%—
FrontierCode53.5%—
LMArena WebDev1625—
FrontierSWE47%—
GSO78.4%—
BigCodeBench Instruct—32.8%
MirrorCode63.9%—
BigCodeBench Complete—40.5%
ALE-Bench2,041—
HumanEval+—62.8%
MBPP+—55.6%

Agentic & Tool Use Claude Fable 5 leads

Claude Fable 5: 54.0 (#2), Llama 3.1-8B: 22.5 (#131)

BenchmarkClaude Fable 5Llama 3.1-8B
APEX-Agents63.6%—
Berkeley Function Calling Leaderboard—25.8%
Remote Labor Index16.1%—
τ²-bench Banking39.7%—
PostTrainBench41.8%—
BALROG—15.1%
GBAEval74.5%—
GDP.pdf30%—
LMArena Search1230—
Vending-Bench 25,680—

Reasoning Claude Fable 5 leads

Claude Fable 5: 76.8 (#6), Llama 3.1-8B: 14.9 (#321)

BenchmarkClaude Fable 5Llama 3.1-8B
CritPt28.6%0%
Chess Puzzles41%0%
LMArena Hard Prompts15081175
DTBench98.4%50.9%
LMCA61.1%5.4%
Epoch Capabilities Index162.06116.57
ARC-AGI-289.2%—
SimpleBench81.9%—
Kagi LLM Benchmark91.4%—
NYT Connections (extended)92.7%—
ARC-AGI-198.5%—
EnigmaEval39.3%—
EBR-Bench39.5%—
Mystery Game Puzzles52%—
Surface Evolver Bench95%—
Bench to the Future 30.13—
PIQA—81.2%

Math Claude Fable 5 leads

Claude Fable 5: 88.5 (#5), Llama 3.1-8B: 10.2 (#317)

BenchmarkClaude Fable 5Llama 3.1-8B
OTIS Mock AIME 2024-2025100%1.7%
LMArena Math15191179
FrontierMath (Tiers 1-3)87%—
FrontierMath Tier 490.2%—
ProofBench95%—
Omni-MATH—13.7%
MATH Level 5—22.9%
FrontierMath Erdős0%—
GSM8K—82.4%

Knowledge Claude Fable 5 leads

Claude Fable 5: 62.2 (#25), Llama 3.1-8B: 8.0 (#307)

BenchmarkClaude Fable 5Llama 3.1-8B
GPQA Diamond85.9%27%
LMArena Expert15341144
SimpleQA Verified70.7%—
MMLU-Pro—40.6%
GPQA (HELM)—24.7%
BoolQ—82.8%
MMLU—56.1%

Multimodal Not comparable

Claude Fable 5: 45.3 (#17), Llama 3.1-8B: —

BenchmarkClaude Fable 5Llama 3.1-8B
LMArena Vision1324—
Blueprint-Bench 238.6%—
Furniture Assembly35.8%—
LMArena Document1496—

Multilingual Claude Fable 5 leads

Claude Fable 5: 57.3 (#9), Llama 3.1-8B: 34.0 (#249)

BenchmarkClaude Fable 5Llama 3.1-8B
LMArena Non-English14811148
LMArena Chinese15431151
LMArena French15051177
LMArena German14861144
LMArena Japanese15061061
LMArena Korean14881053
LMArena Russian15041158
LMArena Spanish14981169

Instruction Following Claude Fable 5 leads

Claude Fable 5: 78.6 (#8), Llama 3.1-8B: 58.9 (#258)

BenchmarkClaude Fable 5Llama 3.1-8B
LMArena Instruction Following15021159
IFEval—74.3%

Long Context Claude Fable 5 leads

Claude Fable 5: 46.3 (#23), Llama 3.1-8B: 35.8 (#238)

BenchmarkClaude Fable 5Llama 3.1-8B
LMArena Longer Query15091182

Writing & Preference Claude Fable 5 leads

Claude Fable 5: 75.9 (#5), Llama 3.1-8B: 29.7 (#290)

BenchmarkClaude Fable 5Llama 3.1-8B
LMArena Text14911187
LMArena Creative Writing14941154
EQ-Bench Creative Writing1943713
LMArena Multi-Turn15041172
WildBench—68.7%
EQ-Bench 41340—

Frequently asked questions

Is Claude Fable 5 better than Llama 3.1-8B?

Claude Fable 5 is the stronger model overall, scoring 66.8 to 23.0 on the Noometry Index. Llama 3.1-8B costs 348× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Which is cheaper, Claude Fable 5 or Llama 3.1-8B?

Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Claude Fable 5 lists at $10 and $50.

Is Claude Fable 5 or Llama 3.1-8B better for coding?

Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 20.2 in the Noometry coding category.

Which has the bigger context window?

Claude Fable 5 does, with 1M tokens against 128K.

How many benchmarks do Claude Fable 5 and Llama 3.1-8B share?

27 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and Llama 3.1-8B has 43.

Related comparisons

Go deeper