Pairwise benchmark snapshot · Aug 17, 2026

Claude 3.5 Haiku vs Doubao Lite 1.5 benchmark

Name: Claude 3.5 Haiku vs Doubao Lite 1.5 AI benchmark comparison
Creator: CompareLLM
License: https://creativecommons.org/licenses/by/4.0/

This page compares Claude 3.5 Haiku (Anthropic) and Doubao Lite 1.5 (ByteDance) using the latest snapshots we have as of Aug 17, 2026. On preference Elo, Doubao Lite 1.5 is ahead by 136 points (1,360 vs 1,224; lmarena (Aug 17, 2026)). On SWE-bench coding, Doubao Lite 1.5 resolves 46% versus 36.2% (swebench (Aug 17, 2026)). Doubao Lite 1.5 streams faster (180 tok/s) while Doubao Lite 1.5 is the cheaper output token ($0.1/1M). Claude 3.5 Haiku has the larger context window (200k tokens). Every cell below is a dated snapshot from a named public source — we do not invent missing scores, and we do not run SWE-bench ourselves.

Share Analysis:

WhatsApp Telegram X / Post LinkedIn Reddit

Anthropic

Claude 3.5 Haiku

Previous cheap Claude. Haiku 4.5 is the current $1/$5 SKU.

Elo 1,224$4/1M out

ByteDance

Doubao Lite 1.5

ByteDance high-speed lightweight model priced at sub-cent levels.

Elo 1,360$0.1/1M out

Swap or Add Any Model to this Matchup

Instant pairwise benchmark recalculation across 400+ LLMs

Model A

Model B

Quick vs:vs Claude Opus 4.5 vs Claude Opus 4.6 vs Claude Sonnet 4.5 vs Claude 3.7 Sonnet

Interactive Comparison Diagram

Matchup Breakdown

Tug-of-War Matchup Matrix

Category wins across intelligence, speed, and pricing efficiency.

Claude 3.5 Haiku (1)vsDoubao Lite 1.5 (8)

Claude 3.5 Haiku: 1W (11%)Overall: Doubao Lite 1.5Doubao Lite 1.5: 8W (89%)

← Claude 3.5 Haiku1 tiesDoubao Lite 1.5 →

Doubao Lite 1.5 (5/5)

Intelligence & Reasoning

Preference Elo, Coding proficiency, SWE-bench & LiveBench accuracy

Preference Elo

1,224vs1,360

Coding Elo

—vs1,320

SWE-bench

36.2%vs46%

LiveBench

44.8%vs54%

GPQA Diamond

54.1%vs64.2%

Claude 3.5 Haiku: 0WDoubao Lite 1.5: 5W

Doubao Lite 1.5 (1/2)

Speed & Latency

Generation throughput and time to first token responsiveness

Output speed

140 tok/svs180 tok/s

Time to first token

110 msvs110 ms

Claude 3.5 Haiku: 0WDoubao Lite 1.5: 1W

Doubao Lite 1.5 (2/3)

Pricing & Capacity

Cost per million tokens and max context window length

Output price

$4/1Mvs$0.1/1M

Input price

$0.8/1Mvs$0.04/1M

Context window

200kvs33k

Claude 3.5 Haiku: 1WDoubao Lite 1.5: 2W

Capability Matchup Radar

Multi-Dimensional Capability Radar

Each spoke is a skill. Farther from the center is better (0–100th percentile). Tap any dot to inspect details.

Tap any node to inspect

Percentile 0–100

Dimensional AdvantageDoubao Lite 1.5 5/6

Preference Elo

Doubao Lite 1.5

136 pts advantage

Throughput Speed

Doubao Lite 1.5

40 tok/s faster

Price Efficiency

Doubao Lite 1.5

$3.90/1M cheaper

Full Benchmark Score Matrix

—

Elo

High is better

1,224

EloHigh is better

1,360+11%

Code Elo

High is better

—

Code EloHigh is better

1,320

LiveBench

High is better

44.8%

LiveBenchHigh is better

54%+21%

SWE-bench

High is better

36.2%

SWE-benchHigh is better

46%+27%

GPQA

High is better

54.1%

GPQAHigh is better

64.2%+19%

TTFT

Low is better

110 ms

TTFTLow is better

110 ms

Speed

High is better

140 tok/s

SpeedHigh is better

180 tok/s+29%

In $

Low is better

$0.8/1M

In $Low is better

$0.04/1M+1900%

Out $

Low is better

$4/1M

Out $Low is better

$0.1/1M+3900%

Context

High is better

+510%200k

ContextHigh is better

33k

Interactive Simulator

Workload Cost & Savings Calculator

Simulate monthly production API costs based on published $/1M tokens.

Save up to 97% with Doubao Lite 1.5

Monthly Volume50M tokens/mo

Quick Presets:

Token Ratio (In vs Out)70% In / 30% Out

RAG / Search (10% out)Coding / Chat (50% out)

Claude 3.5 Haiku$88.00 / mo

In: $28.00Out: $60.00

Doubao Lite 1.5$2.90 / mo

In: $1.40Out: $1.50

Estimated Cost Delta

Doubao Lite 1.5 is estimated to save $85.10/month ($1,021/year).

Based on 50M tokens (30% gen)

Key Selection Recommendations

1
Doubao Lite 1.5:Pick Doubao Lite 1.5 when repo-level coding accuracy is the constraint.
2
Claude 3.5 Haiku:Pick Claude 3.5 Haiku when time-to-first-token matters more than peak Elo.
3
Doubao Lite 1.5:Pick Doubao Lite 1.5 as the default general assistant.

Recommended Workload Routing

Repo / coding agentsDoubao Lite 1.5

Higher SWE-bench (46%).

High-volume chatDoubao Lite 1.5

Lower output list price ($0.1/1M).

Voice / low-latency UIClaude 3.5 Haiku

Lower TTFT (110 ms).

Long-document RAGClaude 3.5 Haiku

Larger window (200k).

Screenshots / visionClaude 3.5 Haiku

Claude 3.5 Haiku is the side marked multimodal in the catalog.

Target Workload	Recommended Pick	Evaluation Rationale
Repo / coding agents	Doubao Lite 1.5	Higher SWE-bench (46%).
High-volume chat	Doubao Lite 1.5	Lower output list price ($0.1/1M).
Voice / low-latency UI	Claude 3.5 Haiku	Lower TTFT (110 ms).
Long-document RAG	Claude 3.5 Haiku	Larger window (200k).
Screenshots / vision	Claude 3.5 Haiku	Claude 3.5 Haiku is the side marked multimodal in the catalog.

Community Sentiment

Cast Your Matchup Vote

0 total votes

Community Discussions (0)

No comments posted on this matchup yet. Be the first to share an evaluation note!

Share card

The real PNG is generated at /compare/claude-3-5-haiku-vs-doubao-1-5-lite/opengraph-image for crawlers.

PNG

CompareLLM Matrix

Verified head-to-head card

LIVE

Anthropic

Claude 3.5 Haiku

Elo 1,224

LiveBench44.8%

SWE-bench36.2%

Speed140 tok/s

Output cost$4/1M

ByteDance

Doubao Lite 1.5

Elo 1,360

LiveBench54%

SWE-bench46%

Speed180 tok/s

Output cost$0.1/1M

Frequently Asked Questions

Which is better overall, Claude 3.5 Haiku or Doubao Lite 1.5?: Doubao Lite 1.5 has the higher preference Elo in our latest snapshot (1,360). “Better” still depends on coding, price, and latency — see the table.
Which is better at coding, Claude 3.5 Haiku or Doubao Lite 1.5?: Doubao Lite 1.5 leads SWE-bench at 46% vs 36.2%. SWE-bench is one harness, not your repo.
Which is cheaper to run in production?: Doubao Lite 1.5 output tokens are $0.1/1M versus $4/1M. Input prices and retry rates still move the real bill.
Which is faster, Claude 3.5 Haiku or Doubao Lite 1.5?: Claude 3.5 Haiku has the lower time-to-first-token (110 ms vs 110 ms). Tokens/sec is a separate column if you care about long completions.
Which has the larger context window?: Claude 3.5 Haiku accepts 200k tokens versus 33k.
Can I self-host either model?: Neither model is marked open-weights here. You are comparing hosted APIs.
Are these scores from CompareLLM’s own evals?: No. V1 aggregates public leaderboards (and optional first-party latency pings). Each cell names its source and date. Read /methodology.
What does preference Elo mean on this page?: Preference Elo is a crowd vote from LMArena / Arena. People see two hidden answers and pick the one they like more. The model that wins more often gets a higher Elo. That means people preferred it — not that it passed a school test. It is not SWE-bench, not accuracy, and not a number we invent. Claude 3.5 Haiku and Doubao Lite 1.5 Elo cells are dated snapshots from the named source — we do not compute Arena ourselves.

Similar Head-to-Head Comparisons

All Claude 3.5 Haiku matchups →|All Doubao Lite 1.5 matchups →

Claude 3.5 Haiku vs Claude Haiku 4.5 (next haiku)Doubao Lite 1.5 vs Doubao Pro 1.5 (next seed)Claude Opus 4.5 vs Doubao Lite 1.5 Claude Opus 4.6 vs Doubao Lite 1.5 Claude Sonnet 4.5 vs Doubao Lite 1.5 Claude Opus 5 vs Doubao Lite 1.5

Claude 3.5 Haiku vs Doubao Lite 1.5 benchmark

Interactive Comparison Diagram

Matchup Breakdown

Tug-of-War Matchup Matrix

Category wins across intelligence, speed, and pricing efficiency.

Claude 3.5 Haiku (1)vsDoubao Lite 1.5 (8)

Claude 3.5 Haiku: 1W (11%)Overall: Doubao Lite 1.5Doubao Lite 1.5: 8W (89%)

← Claude 3.5 Haiku1 tiesDoubao Lite 1.5 →

Doubao Lite 1.5 (5/5)

Intelligence & Reasoning

Preference Elo, Coding proficiency, SWE-bench & LiveBench accuracy

Preference Elo

1,224vs1,360

Coding Elo

—vs1,320

SWE-bench

36.2%vs46%

LiveBench

44.8%vs54%

GPQA Diamond

54.1%vs64.2%

Claude 3.5 Haiku: 0WDoubao Lite 1.5: 5W

Doubao Lite 1.5 (1/2)

Speed & Latency

Generation throughput and time to first token responsiveness

Output speed

140 tok/svs180 tok/s

Time to first token

110 msvs110 ms

Claude 3.5 Haiku: 0WDoubao Lite 1.5: 1W

Doubao Lite 1.5 (2/3)

Pricing & Capacity

Cost per million tokens and max context window length

Output price

$4/1Mvs$0.1/1M

Input price

$0.8/1Mvs$0.04/1M

Context window

200kvs33k

Claude 3.5 Haiku: 1WDoubao Lite 1.5: 2W

Recommended Workload Routing

Repo / coding agentsDoubao Lite 1.5

Higher SWE-bench (46%).

High-volume chatDoubao Lite 1.5

Lower output list price ($0.1/1M).

Voice / low-latency UIClaude 3.5 Haiku

Lower TTFT (110 ms).

Long-document RAGClaude 3.5 Haiku

Larger window (200k).

Screenshots / visionClaude 3.5 Haiku

Claude 3.5 Haiku is the side marked multimodal in the catalog.

Target Workload	Recommended Pick	Evaluation Rationale
Repo / coding agents	Doubao Lite 1.5	Higher SWE-bench (46%).
High-volume chat	Doubao Lite 1.5	Lower output list price ($0.1/1M).
Voice / low-latency UI	Claude 3.5 Haiku	Lower TTFT (110 ms).
Long-document RAG	Claude 3.5 Haiku	Larger window (200k).
Screenshots / vision	Claude 3.5 Haiku	Claude 3.5 Haiku is the side marked multimodal in the catalog.

Frequently Asked Questions

Which is better overall, Claude 3.5 Haiku or Doubao Lite 1.5?

Doubao Lite 1.5 has the higher preference Elo in our latest snapshot (1,360). “Better” still depends on coding, price, and latency — see the table.

Which is better at coding, Claude 3.5 Haiku or Doubao Lite 1.5?

Doubao Lite 1.5 leads SWE-bench at 46% vs 36.2%. SWE-bench is one harness, not your repo.

Which is cheaper to run in production?

Doubao Lite 1.5 output tokens are $0.1/1M versus $4/1M. Input prices and retry rates still move the real bill.

Which is faster, Claude 3.5 Haiku or Doubao Lite 1.5?

Claude 3.5 Haiku has the lower time-to-first-token (110 ms vs 110 ms). Tokens/sec is a separate column if you care about long completions.

Which has the larger context window?

Claude 3.5 Haiku accepts 200k tokens versus 33k.

Can I self-host either model?

Neither model is marked open-weights here. You are comparing hosted APIs.

Are these scores from CompareLLM’s own evals?

No. V1 aggregates public leaderboards (and optional first-party latency pings). Each cell names its source and date. Read /methodology.

What does preference Elo mean on this page?

Preference Elo is a crowd vote from LMArena / Arena. People see two hidden answers and pick the one they like more. The model that wins more often gets a higher Elo. That means people preferred it — not that it passed a school test. It is not SWE-bench, not accuracy, and not a number we invent. Claude 3.5 Haiku and Doubao Lite 1.5 Elo cells are dated snapshots from the named source — we do not compute Arena ourselves.