CompareLLM
CompareLLM.ai
Live
Explore
Compare
Decide
Track
Learn
…
CompareLLM
CompareLLM.ai
Precision Benchmarks

Programmatic, dated AI model benchmarks, head-to-head comparisons, and Stack Engine presets.

Daily ingest · 06:00 UTC

Analytics & Benchmarks

  • AI Model Leaderboard
  • Head-to-Head Compare Hub
  • Models Directory
  • Stack Engine Presets
  • Frontier Models
  • Open Weights Catalog

Guides & Intent Lists

  • Best LLM Lists (2026)
  • Best Coding LLM
  • Best Cheap LLM
  • Fastest Low-Latency LLM
  • Claude vs GPT Benchmark
  • What is Elo?
  • Methodology Guides
  • News & Dispatches

Transparency & API

  • Evaluation Methodology
  • Benchmark Changelog
  • Public JSON API
  • llms.txt Specification
  • Privacy Policy
  • Sign In / Account

© 2026 CompareLLM. Public benchmark data aggregated from Arena Elo, LiveBench, SWE-bench & OpenRouter.

Every score has a dated snapshot.

Theme:
Currency:
  1. Home
  2. Comparisons
  3. Claude 3.5 Haiku vs Anthropic: Claude Sonnet 4.6

Pairwise benchmark snapshot · Aug 17, 2026noindex (thin pair)

Claude 3.5 Haiku vs Anthropic: Claude Sonnet 4.6 benchmark

In this head-to-head showdown, Claude 3.5 Haiku is more budget-friendly at $4/1M per 1M output tokens. Review the complete breakdown below to determine which model best fits your performance and budget requirements.

Executive Comparison Verdict

Bottom Line: Both Claude 3.5 Haiku and Anthropic: Claude Sonnet 4.6 present distinct architectural strengths. Compare the verified benchmark matrix below to choose the model tailored to your specific application requirements.
Share Analysis:Share:
WhatsAppTelegramXLinkedIn
Executive Verdict Summary

Claude 3.5 Haiku holds empirical advantage (2 of 3 metrics)

Coding / SWESWE-bench
Claude 3.5 HaikuTop code solve rate
ReasoningGPQA / Live
Claude 3.5 HaikuTop logical accuracy
Best Value$/1M Tok
Claude 3.5 HaikuLowest API billing cost
Speed / TokTok/Sec
Claude 3.5 HaikuFastest stream rate

Verified Advantage Breakdown

3 benchmarks evaluated across capability, speed, and cost

Claude 3.5 Haiku (2)Anthropic: Claude Sonnet 4.6 (1)
Claude 3.5 Haiku
2 of 3 Wins
✓Input price✓Output price
Anthropic: Claude Sonnet 4.6
1 of 3 Wins
✓Context window
Anthropic
Leads 2 of 3
Claude 3.5 Haiku

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

Elo 1,224In: $0.8 · Out: $4/1M
Anthropic
Leads 1 of 3
Anthropic: Claude Sonnet 4.6

Auto-discovered from OpenRouter (anthropic/claude-sonnet-4.6). Preview until a second source matches.

In: $3 · Out: $15/1M
Top Rival Showdowns for Claude 3.5 Haiku
All Matchups
Compare vs:vs GPT-5vs Gemini 3 Provs Grok 4.6vs DeepSeek V4 Provs Qwen 3 Max
Preference Leader

No shared data

Single model data

Throughput Leader

No shared data

No latency data

Value per Dollar LeaderClaude 3.5 Haiku Wins

Claude 3.5 Haiku

$4 / 1M output

Capability Percentiles

Relative percentile scores computed across all active models in the benchmark catalog.

Multi-Dimensional Capability Radar

Leaders Matchup Capability Radar

Comparing top standard models across 6 key capability dimensions. Tap any spoke or dot to inspect.

Tap any node to inspect
0–100 %ile

Overall Matchup Breakdown

Category wins across reasoning intelligence, generation speed, and token cost.

Claude 3.5 Haiku (8)vsAnthropic: Claude Sonnet 4.6 (1)
Claude 3.5 Haiku: 8W (89%)Overall: Claude 3.5 HaikuAnthropic: Claude Sonnet 4.6: 1W (11%)
← Claude 3.5 HaikuAnthropic: Claude Sonnet 4.6 →
🏆Claude 3.5 Haiku Leads(4/5)

Intelligence & Reasoning

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

Benchmark
Claude 3.5 HaikuvsAnthropic: Claude Sonnet 4.6
Preference Elo
1,224vs—
Coding Elo
—vs—
SWE-bench
36.2%vs—
LiveBench
44.8%vs—
GPQA Diamond
54.1%vs—
Claude 3.5 Haiku: 4WAnthropic: Claude Sonnet 4.6: 0W
🏆Claude 3.5 Haiku Leads(2/2)

Speed & Latency

Generation throughput and time to first token responsiveness

Benchmark
Claude 3.5 HaikuvsAnthropic: Claude Sonnet 4.6
Output speed
140 tok/svs—
Time to first token
110 msvs—
Claude 3.5 Haiku: 2WAnthropic: Claude Sonnet 4.6: 0W
🏆Claude 3.5 Haiku Leads(2/3)

Pricing & Capacity

Cost per million tokens and max context window length

Benchmark
Claude 3.5 HaikuvsAnthropic: Claude Sonnet 4.6
Output price
$4/1Mvs$15/1M+$11/1M
Input price
$0.8/1Mvs$3/1M+$2.2/1M
Context window
200kvs1M+800k
Claude 3.5 Haiku: 2WAnthropic: Claude Sonnet 4.6: 1W

Side-by-Side Benchmark Matrix

Claude 3.5 HaikuAnthropic: Claude Sonnet 4.6
Preference Elo
Claude 3.5 Haiku1,224
Anthropic: Claude Sonnet 4.6—
LiveBench
Claude 3.5 Haiku44.8%
Anthropic: Claude Sonnet 4.6—
SWE-bench
Claude 3.5 Haiku36.2%
Anthropic: Claude Sonnet 4.6—
GPQA Diamond
Claude 3.5 Haiku54.1%
Anthropic: Claude Sonnet 4.6—
Time to first token
Claude 3.5 Haiku110 ms
Anthropic: Claude Sonnet 4.6—
Output speed
Claude 3.5 Haiku140 tok/s
Anthropic: Claude Sonnet 4.6—
Input priceClaude 3.5 Haiku +$2.2/1M
Claude 3.5 Haiku$0.8/1M
Anthropic: Claude Sonnet 4.6$3/1M
Output priceClaude 3.5 Haiku +$11/1M
Claude 3.5 Haiku$4/1M
Anthropic: Claude Sonnet 4.6$15/1M
Context windowAnthropic: Claude Sonnet 4.6 +800k
Claude 3.5 Haiku200k
Anthropic: Claude Sonnet 4.61M
BenchmarkClaude 3.5 HaikuAnthropic: Claude Sonnet 4.6Advantage Delta
Preference Elo
1,224
lmarena · Aug 17, 2026
——
LiveBench
44.8%
livebench · Aug 17, 2026
——
SWE-bench
36.2%
swebench · Aug 17, 2026
——
GPQA Diamond
54.1%
seed-bootstrap · Aug 1, 2026
——
Time to first token
110 ms
seed-bootstrap · Aug 1, 2026
——
Output speed
140 tok/s
seed-bootstrap · Aug 1, 2026
——
Input price
$0.8/1M
seed-bootstrap · Aug 1, 2026
$3/1M
openrouter · Aug 17, 2026
+$2.2/1M
Output price
$4/1M
seed-bootstrap · Aug 1, 2026
$15/1M
openrouter · Aug 17, 2026
+$11/1M
Context window
200k
seed-bootstrap · Aug 1, 2026
1M
openrouter · Aug 17, 2026
+800k

Workload Cost & Savings Calculator

Simulate monthly production API costs in USD (US Dollar).

Save 73% with Claude 3.5 Haiku
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: $28Out: $60
Anthropic: Claude Sonnet 4.6$330.00 / mo
In: $105Out: $225
Estimated Cost Delta

Claude 3.5 Haiku is estimated to save $242.00/month ($2,904/year).

Based on 50M tokens (30% gen)

Key Selection Recommendations

  • 1
    Claude 3.5 Haiku:Pick Claude 3.5 Haiku when you are optimizing output cost.
  • 2
    Anthropic: Claude Sonnet 4.6:Pick Anthropic: Claude Sonnet 4.6 for million-token RAG or long-document jobs.

Recommended Workload Routing

Repo / coding agentsinsufficient data

Need SWE-bench on both sides.

High-volume chatClaude 3.5 Haiku

Lower output list price ($4/1M).

Voice / low-latency UIinsufficient data

Need TTFT on both sides.

Long-document RAGAnthropic: Claude Sonnet 4.6

Larger window (1M).

Screenshots / visionClaude 3.5 Haiku

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

Target WorkloadRecommended PickEvaluation Rationale
Repo / coding agentsinsufficient dataNeed SWE-bench on both sides.
High-volume chatClaude 3.5 HaikuLower output list price ($4/1M).
Voice / low-latency UIinsufficient dataNeed TTFT on both sides.
Long-document RAGAnthropic: Claude Sonnet 4.6Larger window (1M).
Screenshots / visionClaude 3.5 HaikuClaude 3.5 Haiku is the side marked multimodal in the catalog.

Cast Your Matchup Vote

0 total votes

Community Discussions (0)

Sign in to cast your verified vote, bookmark models, and participate in benchmark discussions.Sign In to Post

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

Social Share Card Preview

Dynamic OpenGraph banner generated at /compare/claude-3-5-haiku-vs-claude-sonnet-4-6/opengraph-image

Save PNG

CompareLLM AI Matrix

Verified Head-to-Head Telemetry

LIVE BENCHMARK
VS
Anthropic

Claude 3.5 Haiku

Elo 1,224
SWE-bench36.2%
Speed (tok/s)140 tok/s
Input / 1M Tokens$0.8/1M
Output / 1M Tokens$4/1M
Anthropic

Anthropic: Claude Sonnet 4.6

Elo n/a
SWE-bench—
Speed (tok/s)1M
Input / 1M Tokens$3/1M
Output / 1M Tokens$15/1M

Frequently Asked Questions

Which is better overall, Claude 3.5 Haiku or Anthropic: Claude Sonnet 4.6?
We do not have preference Elo for both models, so we do not declare an overall winner. Compare the metrics that exist.
Which is better at coding, Claude 3.5 Haiku or Anthropic: Claude Sonnet 4.6?
SWE-bench is missing for at least one model, so we do not rank coding from a single number.
Which is cheaper to run in production?
Claude 3.5 Haiku output tokens are $4/1M versus $15/1M. Input prices and retry rates still move the real bill.
Which is faster, Claude 3.5 Haiku or Anthropic: Claude Sonnet 4.6?
We do not have TTFT for both models.
Which has the larger context window?
Anthropic: Claude Sonnet 4.6 accepts 1M tokens versus 200k.
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 Anthropic: Claude Sonnet 4.6 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 Anthropic: Claude Sonnet 4.6 matchups →
Claude 3.5 Haiku vs Claude Haiku 4.5 (next haiku)Claude Opus 4.5 vs Claude 3.5 HaikuClaude Opus 4.6 vs Claude 3.5 HaikuClaude Sonnet 4.5 vs Claude 3.5 HaikuClaude Opus 5 vs Claude 3.5 HaikuClaude Fable 5 vs Claude 3.5 Haiku