Enter opens the highlighted result. Esc closes.

Nova Lite 1.0 vs Claude Haiku 4.5

Claude Haiku 4.5 leads Nova Lite 1.0 on the Artificial Analysis Intelligence Index, 16.9 to 6.7, and Nova Lite 1.0 costs 20.8x less per output token ($0.24 vs $5.00 per million). Updated October 22, 2026; 28 benchmarks have a result for both.

Replace Nova Lite 1.0 Replace Claude Haiku 4.5
Nova Lite 1.0Amazon
Coding
No data
Overall
No data
Output $/M
$0.24
0
Claude Haiku 4.5Anthropic
Coding
35.6
Overall
41.0
Output $/M
$5.00
28
vs boards won of 28 shared

Nova Lite 1.0 or Claude Haiku 4.5: which to pick

Only Claude Haiku 4.5 has a coding score, 35.6 on the Vibe Coding Index; Nova Lite 1.0 has none, so the two cannot be ranked on coding.

Pick Nova Lite 1.0 for output tokens 20.8x cheaper ($0.24 vs $5.00 per million) and a larger context window (300K against 200K tokens).

Model ANova Lite 1.0AmazonNot benchmarkedNo coding benchmarkNo cost per solved taskModel BClaude Haiku 4.5Anthropic35.6/1004 independent benchmarks$0.56 per solved task
Shared results
28
Nova Lite 1.0 only
13
Claude Haiku 4.5 only
67
Like-for-like categories
6 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
Nova Lite 1.0 not benchmarked, Claude Haiku 4.5 35.6 from 4 benchmarks.
Widest category
Reasoning, 36.2 points apart on average over 3 shared benchmarks.
Widest single gap
LMArena Text: expert: Nova Lite 1.0 1201 against Claude Haiku 4.5 1443, 40.3 points apart on a 0 to 100 scale.
Output price
Nova Lite 1.0 $0.24 against Claude Haiku 4.5 $5.00 per million output tokens, 20.8x apart.
Context window
Nova Lite 1.0 takes 300K tokens, Claude Haiku 4.5 200K.

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
CategoryNova Lite 1.0BarsClaude Haiku 4.5Ties
Coding02
Intelligence01
Reasoning03
Math02
Knowledge04
General016
All shared boards028
CodingLike-for-likeSix of these boards feed the Vibe Coding Index02
Coding benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
SWE-bench VerifiedNot measured66.6%Only Claude Haiku 4.5 measured
SWE-bench MultilingualNot measured64.7%Only Claude Haiku 4.5 measured
LiveBench codingNot measured48.1%Only Claude Haiku 4.5 measured
LMArena WebDevNot measured1330Only Claude Haiku 4.5 measured
AA Coding IndexNot measured43.9Only Claude Haiku 4.5 measured
LiveCodeBench (Vals AI)Not measured41.2%Only Claude Haiku 4.5 measured
SWE-bench Verified (Vals AI)Not measured66.6%Only Claude Haiku 4.5 measured
Vibe Code Bench (Vals AI)Not measured11.4%Only Claude Haiku 4.5 measured
Code migration (Vals AI)Not measured7.5%Only Claude Haiku 4.5 measured
ProgramBench (Vals AI)Not measured0.0%Only Claude Haiku 4.5 measured
Terminal-Bench 2.1 (Vals AI)Not measured43.8%Only Claude Haiku 4.5 measured
LiveCodeBench (Artificial Analysis run)16.7%Not measuredOnly Nova Lite 1.0 measured
Terminal-Bench Hard (Artificial Analysis run)0.8%Not measuredOnly Nova Lite 1.0 measured
Terminal-BenchNot measured35.5%Only Claude Haiku 4.5 measured
WeirdMLNot measured45.4%Only Claude Haiku 4.5 measured
LMArena Text: coding12391454Claude Haiku 4.5
LMArena Text: software and IT12411439Claude Haiku 4.5
LMArena WebDev: ReactNot measured1326Only Claude Haiku 4.5 measured
LMArena WebDev: HTMLNot measured1318Only Claude Haiku 4.5 measured
Neither model has a result on the other 20 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
AA Intelligence Index6.716.9Claude Haiku 4.5
Epoch Capabilities IndexNot measured142.2Only Claude Haiku 4.5 measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index13 boards, none shared
Agentic work benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
AA Agentic IndexNot measured8.0Only Claude Haiku 4.5 measured
LiveBench researchNot measured42.2%Only Claude Haiku 4.5 measured
LiveBench browserNot measured40.0%Only Claude Haiku 4.5 measured
LiveBench long agenticNot measured11.1%Only Claude Haiku 4.5 measured
Finance Agent (Vals AI)Not measured31.0%Only Claude Haiku 4.5 measured
Tax Agent Bench (Vals AI)Not measured5.9%Only Claude Haiku 4.5 measured
Legal Research Bench (Vals AI)Not measured10.6%Only Claude Haiku 4.5 measured
Excel Modeling Benchmark (Vals AI)Not measured22.7%Only Claude Haiku 4.5 measured
τ²-Bench Telecom17.5%Not measuredOnly Nova Lite 1.0 measured
BALROGNot measured31.2%Only Claude Haiku 4.5 measured
DeepResearch BenchNot measured45.5%Only Claude Haiku 4.5 measured
Vending-Bench 2Not measured$459Only Claude Haiku 4.5 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured68.7%Only Claude Haiku 4.5 measured
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index03
Reasoning benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
GPQA DiamondNot measured72.2%Only Claude Haiku 4.5 measured
LiveBench reasoningNot measured61.7%Only Claude Haiku 4.5 measured
MMMU Pro (Vals AI)Not measured46.1%Only Claude Haiku 4.5 measured
GPQA Diamond (Artificial Analysis run)43.3%Not measuredOnly Nova Lite 1.0 measured
Humanity's Last Exam (Artificial Analysis run)4.3%Not measuredOnly Nova Lite 1.0 measured
GPQA Diamond (Epoch AI run)Not measured71.2%Only Claude Haiku 4.5 measured
ARC-AGINot measured47.7%Only Claude Haiku 4.5 measured
ARC-AGI-2Not measured4.0%Only Claude Haiku 4.5 measured
Chess puzzlesNot measured8.0%Only Claude Haiku 4.5 measured
LMArena Text: hard prompts12201421Claude Haiku 4.5
LMArena Text: expert12011443Claude Haiku 4.5
LMArena Text: hard prompts in English12261435Claude Haiku 4.5
Neither model has a result on the other 10 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
AIMENot measured82.7%Only Claude Haiku 4.5 measured
MGSM (Vals AI)Not measured92.2%Only Claude Haiku 4.5 measured
AA Math Index7.0Not measuredOnly Nova Lite 1.0 measured
AIME 2025 (Artificial Analysis run)7.0%Not measuredOnly Nova Lite 1.0 measured
OTIS Mock AIMENot measured66.7%Only Claude Haiku 4.5 measured
MATH level 5Not measured96.4%Only Claude Haiku 4.5 measured
LMArena Text: math12271398Claude Haiku 4.5
LMArena Text: mathematical work12351430Claude Haiku 4.5
Neither model has a result on the other 5 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
MMLU ProNot measured78.7%Only Claude Haiku 4.5 measured
LegalBench (Vals AI)Not measured81.2%Only Claude Haiku 4.5 measured
CaseLaw (Vals AI)Not measured56.5%Only Claude Haiku 4.5 measured
CorpFin (Vals AI)Not measured60.6%Only Claude Haiku 4.5 measured
TaxEval (Vals AI)Not measured67.5%Only Claude Haiku 4.5 measured
MortgageTax (Vals AI)Not measured62.2%Only Claude Haiku 4.5 measured
MedQA (Vals AI)Not measured79.6%Only Claude Haiku 4.5 measured
MedCode (Vals AI)Not measured32.7%Only Claude Haiku 4.5 measured
MMLU Pro (Artificial Analysis run)59.0%Not measuredOnly Nova Lite 1.0 measured
SimpleQA VerifiedNot measured13.2%Only Claude Haiku 4.5 measured
MMLU77.0%Not measuredOnly Nova Lite 1.0 measured
LMArena Text: science12321406Claude Haiku 4.5
LMArena Text: medicine and healthcare12131394Claude Haiku 4.5
LMArena Text: legal and government12381400Claude Haiku 4.5
LMArena Text: business and finance12191400Claude Haiku 4.5
Neither model has a result on the other 3 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Web design benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
Design Arena: websitesNot measured1122Only Claude Haiku 4.5 measured
Design Arena: UI componentsNot measured1097Only Claude Haiku 4.5 measured
Design Arena: game devNot measured1097Only Claude Haiku 4.5 measured
Design Arena: data vizNot measured1124Only Claude Haiku 4.5 measured
Design Arena: 3DNot measured1071Only Claude Haiku 4.5 measured
Design Arena: all code categoriesNot measured1114Only Claude Haiku 4.5 measured
Design Arena: SVGNot measured1025Only Claude Haiku 4.5 measured
Design Arena: ASCII artNot measured1140Only Claude Haiku 4.5 measured
LiveBench designNot measured59.9%Only Claude Haiku 4.5 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index016
General benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
LiveBench data analysisNot measured59.3%Only Claude Haiku 4.5 measured
Public Benefits Bench (Vals AI)Not measured54.3%Only Claude Haiku 4.5 measured
SAGE (Vals AI)Not measured31.8%Only Claude Haiku 4.5 measured
MedScribe (Vals AI)Not measured85.2%Only Claude Haiku 4.5 measured
Vals Multimodal IndexNot measured42.9%Only Claude Haiku 4.5 measured
IFBench34.1%Not measuredOnly Nova Lite 1.0 measured
Long context reasoning (AA-LCR)19.0%Not measuredOnly Nova Lite 1.0 measured
DTBenchNot measured73.6%Only Claude Haiku 4.5 measured
LMCANot measured30.9%Only Claude Haiku 4.5 measured
ExploitBenchNot measured13.7%Only Claude Haiku 4.5 measured
LMArena Text12291397Claude Haiku 4.5
LMArena Text: instruction following12051414Claude Haiku 4.5
LMArena Text: creative writing11971371Claude Haiku 4.5
LMArena Vision990Not measuredOnly Nova Lite 1.0 measured
LMArena Text: English12431414Claude Haiku 4.5
LMArena Text: non-English12081377Claude Haiku 4.5
LMArena Text: Chinese12251418Claude Haiku 4.5
LMArena Text: German12291374Claude Haiku 4.5
LMArena Text: Japanese11531338Claude Haiku 4.5
LMArena Text: Korean11531346Claude Haiku 4.5
LMArena Text: French12381411Claude Haiku 4.5
LMArena Text: Spanish12261421Claude Haiku 4.5
LMArena Text: Russian12161381Claude Haiku 4.5
LMArena Text: PolishNot measured1343Only Claude Haiku 4.5 measured
LMArena Text: multi-turn11991409Claude Haiku 4.5
LMArena Text: longer queries12341427Claude Haiku 4.5
LMArena Text: writing and language12111386Claude Haiku 4.5
LMArena Text: entertainment and media11761369Claude Haiku 4.5
LMArena Vision: English999Not measuredOnly Nova Lite 1.0 measured
LMArena DocumentNot measured1420Only Claude Haiku 4.5 measured
Neither model has a result on the other 16 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Nova Lite 1.0 and Claude Haiku 4.5
BenchmarkNova Lite 1.0Claude Haiku 4.5Leader
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

Nova Lite 1.0$0.0002Fits in one request
Claude Haiku 4.5$0.0035Fits in one request

Nova Lite 1.0 costs less

Repository review

50K input + 3K output tokens

Nova Lite 1.0$0.0037Fits in one request
Claude Haiku 4.5$0.065Fits in one request

Nova Lite 1.0 costs less

Agent loop

200K input + 10K output tokens

Nova Lite 1.0$0.014Fits in one request
Claude Haiku 4.5$0.250Exceeds the context window

Nova Lite 1.0 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 Nova Lite 1.0 and Claude Haiku 4.5
SpecificationAmazon Nova Lite 1.0Anthropic Claude Haiku 4.5
MakerAmazonAnthropic
WeightsNo public weights listedNo public weights listed
ReleasedDecember 5, 2024October 15, 2025
Context window300K200K
Input $/Mtok$0.06$1.00
Output $/Mtok$0.24$5.00
Coding indexNot benchmarked35.6
Independent measurements behind it04
Cost per solved taskNot computable$0.56
SourceOpenRouter listingOpenRouter listing

Questions

Is Nova Lite 1.0 better than Claude Haiku 4.5 for coding?

Only Claude Haiku 4.5 has a coding score, 35.6 on the Vibe Coding Index; Nova Lite 1.0 has none, so the two cannot be ranked on coding. Pick Nova Lite 1.0 for output tokens 20.8x cheaper ($0.24 vs $5.00 per million) and a larger context window (300K against 200K tokens).

Which is cheaper, Nova Lite 1.0 or Claude Haiku 4.5?

Nova Lite 1.0 costs $0.06 per million input tokens and $0.24 per million output tokens; Claude Haiku 4.5 costs $1.00 and $5.00. Cost per solved task is known for Claude Haiku 4.5 only.

Which has the bigger context window, Nova Lite 1.0 or Claude Haiku 4.5?

Nova Lite 1.0, with 300K tokens against 200K.

How many benchmarks is the Nova Lite 1.0 vs Claude Haiku 4.5 comparison based on?

28 benchmarks have a result for both models as of October 22, 2026. Nova Lite 1.0's coding index rests on 0 independent benchmarks and Claude Haiku 4.5's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next