Enter opens the highlighted result. Esc closes.

Granite 4.2 8B vs Llama 3.2 3B Instruct

Granite 4.2 8B leads Llama 3.2 3B Instruct on the Artificial Analysis Intelligence Index, 11.1 to 5.7, and Granite 4.2 8B costs 1.7x less per output token ($0.20 vs $0.3325 per million). Updated October 22, 2026; 24 benchmarks have a result for both.

Replace Granite 4.2 8B Replace Llama 3.2 3B Instruct
Granite 4.2 8BIBM Granite
Coding
No data
Overall
No data
Output $/M
$0.20
24
Llama 3.2 3B InstructMeta Llama
Coding
No data
Overall
No data
Output $/M
$0.3325
0
vs boards won of 24 shared

Granite 4.2 8B or Llama 3.2 3B Instruct: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick Granite 4.2 8B for output tokens 1.7x cheaper ($0.20 vs $0.3325 per million).

Model AGranite 4.2 8BIBM GraniteNot benchmarkedNo coding benchmarkNo cost per solved taskModel BLlama 3.2 3B InstructMeta LlamaNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
24
Granite 4.2 8B only
3
Llama 3.2 3B Instruct only
11
Like-for-like categories
5 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Granite 4.2 8B not benchmarked, Llama 3.2 3B Instruct not benchmarked.
Widest category
Coding, 45.4 points apart on average over 2 shared benchmarks.
Widest single gap
LMArena Text: Chinese: Granite 4.2 8B 1367 against Llama 3.2 3B Instruct 1017, 58.3 points apart on a 0 to 100 scale.
Output price
Granite 4.2 8B $0.20 against Llama 3.2 3B Instruct $0.3325 per million output tokens, 1.7x apart.
Context window
Both take 131K tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGranite 4.2 8BBarsLlama 3.2 3B InstructTies
Coding20
Intelligence10
Reasoning50
Knowledge40
General120
All shared boards240
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
Coding benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
AA Coding Index22.4Not measuredOnly Granite 4.2 8B measured
LiveCodeBench (Artificial Analysis run)Not measured8.3%Only Llama 3.2 3B Instruct measured
SciCode31.5%Not measuredOnly Granite 4.2 8B measured
LMArena Text: coding13781098Granite 4.2 8B
LMArena Text: software and IT13671103Granite 4.2 8B
Neither model has a result on the other 34 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
AA Intelligence Index11.15.7Granite 4.2 8B
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Agentic work benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
AA Agentic Index1.3Not measuredOnly Granite 4.2 8B measured
τ²-Bench TelecomNot measured21.1%Only Llama 3.2 3B Instruct measured
BALROGNot measured10.1%Only Llama 3.2 3B Instruct measured
Berkeley Function Calling Leaderboard (BFCL)Not measured21.9%Only Llama 3.2 3B Instruct measured
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
GPQA Diamond (Artificial Analysis run)63.1%25.5%Granite 4.2 8B
Humanity's Last Exam (Artificial Analysis run)9.7%5.3%Granite 4.2 8B
LMArena Text: hard prompts13271095Granite 4.2 8B
LMArena Text: expert13861090Granite 4.2 8B
LMArena Text: hard prompts in English13451141Granite 4.2 8B
Neither model has a result on the other 17 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Math benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
AA Math IndexNot measured3.3Only Llama 3.2 3B Instruct measured
AIME 2025 (Artificial Analysis run)Not measured3.3%Only Llama 3.2 3B Instruct measured
LMArena Text: mathNot measured1126Only Llama 3.2 3B Instruct measured
LMArena Text: mathematical workNot measured1128Only Llama 3.2 3B Instruct measured
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
MMLU Pro (Artificial Analysis run)Not measured34.7%Only Llama 3.2 3B Instruct measured
LMArena Text: science13361106Granite 4.2 8B
LMArena Text: medicine and healthcare13361104Granite 4.2 8B
LMArena Text: legal and government13271115Granite 4.2 8B
LMArena Text: business and finance13221074Granite 4.2 8B
Neither model has a result on the other 13 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index120
General benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
IFBenchNot measured26.2%Only Llama 3.2 3B Instruct measured
Long context reasoning (AA-LCR)45.0%4.3%Granite 4.2 8B
LMArena Text13191110Granite 4.2 8B
LMArena Text: instruction following13001089Granite 4.2 8B
LMArena Text: creative writing12361094Granite 4.2 8B
LMArena Text: English13371180Granite 4.2 8B
LMArena Text: non-English13011019Granite 4.2 8B
LMArena Text: Chinese13671017Granite 4.2 8B
LMArena Text: GermanNot measured1056Only Llama 3.2 3B Instruct measured
LMArena Text: Russian1282949Granite 4.2 8B
LMArena Text: multi-turn13011105Granite 4.2 8B
LMArena Text: longer queries13251100Granite 4.2 8B
LMArena Text: writing and language12691096Granite 4.2 8B
LMArena Text: entertainment and media12711083Granite 4.2 8B
Neither model has a result on the other 32 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Granite 4.2 8B and Llama 3.2 3B Instruct
BenchmarkGranite 4.2 8BLlama 3.2 3B InstructLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Granite 4.2 8B$0.0002Fits in one request
Llama 3.2 3B Instruct$0.0002Fits in one request

Granite 4.2 8B costs less

Repository review

50K input + 3K output tokens

Granite 4.2 8B$0.0046Fits in one request
Llama 3.2 3B Instruct$0.0035Fits in one request

Llama 3.2 3B Instruct costs less

Agent loop

200K input + 10K output tokens

Granite 4.2 8B$0.018Exceeds the context window
Llama 3.2 3B Instruct$0.013Exceeds the context window

Llama 3.2 3B Instruct costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Granite 4.2 8B and Llama 3.2 3B Instruct
SpecificationIBM Granite 4.2 8BMeta Llama 3.2 3B Instruct
MakerIBM GraniteMeta Llama
WeightsOpen weightsOpen weights
ReleasedAugust 31, 2026September 25, 2024
Context window131K131K
Input $/Mtok$0.08$0.0505
Output $/Mtok$0.20$0.3325
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Granite 4.2 8B better than Llama 3.2 3B Instruct for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Granite 4.2 8B for output tokens 1.7x cheaper ($0.20 vs $0.3325 per million).

Which is cheaper, Granite 4.2 8B or Llama 3.2 3B Instruct?

Granite 4.2 8B costs $0.08 per million input tokens and $0.20 per million output tokens; Llama 3.2 3B Instruct costs $0.0505 and $0.3325. Cost per solved task is known for neither of them.

Which has the bigger context window, Granite 4.2 8B or Llama 3.2 3B Instruct?

Neither: both take 131K tokens.

How many benchmarks is the Granite 4.2 8B vs Llama 3.2 3B Instruct comparison based on?

24 benchmarks have a result for both models as of October 22, 2026. Granite 4.2 8B's coding index rests on 0 independent benchmarks and Llama 3.2 3B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next