LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Claude Haiku 5.5

anthropic·anthropic/claude-haiku-5-5·GA·Closed·claude-haiku series·NEW
Fast Claude model for responsive assistance, classification, and lightweight agents

Specs & pricing

Input / output per 1M tokens
Official price·Anthropic
$0.10 / $0.50
Blended $0.20 · Cache read $0.01
Lowest paid·IteraComputeGateway
$0.05 / $0.25
Blended $0.10 · 3.4× spread
1 more $0 channels
Context
1,000,000
Output limit
128,000
Knowledge cutoff
2026-06
Released / updated
2026-10-07 / 2026-10-07
Capabilities
✓ Reasoning✓ Tool use✓ Structured outputTemperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF

Claude Haiku 5.5 by anthropic is offered by 59 providers on this page. Public prices are shown for 57 of them. Its official list price is $0.10 per 1M input tokens and $0.50 per 1M output tokens. The lowest paid channel is IteraCompute at $0.05 / $0.25 per 1M, about 2.0× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.

Artificial Analysis rates it 36.5 on the Intelligence Index, with 43.6 for Finance & Accounting, its strongest domain. Against its lowest blended price of $0.10 per 1M, that is roughly 365 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 230 tokens per second, with 303.69s to the first token. Running one task of the Artificial Analysis suite costs about $0.4110, which reflects how many tokens its reasoning consumes rather than the unit price alone.

The context window is 1,000,000 tokens at the reference host, but hosts report different limits, from 200,000 to 1,000,000, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2026-06.

Quality & performanceArtificial Analysis · Intelligence Index v5 · rep. tier Max

Intelligence36.5
Value365
Engineering42.5
Finance & Accounting43.6
Legal38.7
Healthcare & Medical37.5
Strategy & Ops39.1
Economics42.7
Output speed230 tok/s
TTFT303.69 s
Cost per task$0.41
Value formulaIQ 36.5 ÷ min blended $0.100 = 365
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
LowIQ 18.6 · 154 tok/s
MediumIQ 24.6 · 160 tok/s
HighIQ 29.9 · 155 tok/s
XhighIQ 34.6 · 179 tok/s
MaxIQ 36.5 · 230 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 59 providers57 with public prices · 1 free

  • ProviderInput / Output
  • GitLab DuoCloud
    duo-chat-haiku-5-5
    Free
    Cache read $0Context 1,000,000Output limit 128,000
  • IteraComputeGateway
    $0.05 / $0.25
    tiered >100K: $0.25
    Cache read $0.005Context 1,000,000Output limit 128,000
  • Outer LabsGateway
    $0.06 / $0.30
    tiered >100K: $0.30
    Cache read $0.006Context 1,000,000Output limit 128,000
  • AnthropicOfficialFirst-party
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • Azure Cognitive ServicesCloud
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • Merge GatewayGateway
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • DevPass (LLM Gateway)Gateway
    $0.10 / $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • AixyGateway
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • SynthoraiGateway
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
  • AzureCloud
    $0.10 / $0.50
    tiered >100K: $0.50
    Cache read $0.01Context 1,000,000Output limit 128,000
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
GitLab Duo
duo-chat-haiku-5-5
CloudFree$0$01,000,000128,000
IteraComputeGateway$0.05
tiered >100K: $0.25
$0.25$0.005$0.0631,000,000128,000
Outer LabsGateway$0.06
tiered >100K: $0.30
$0.30$0.006$0.0751,000,000128,000
AnthropicOfficialFirst-party$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000
Azure Cognitive ServicesCloud$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000
Merge GatewayGateway$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000
DevPass (LLM Gateway)Gateway$0.10$0.50$0.01$0.131,000,000128,000
AixyGateway$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000
SynthoraiGateway$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000
AzureCloud$0.10
tiered >100K: $0.50
$0.50$0.01$0.131,000,000128,000

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

Toggle (on / off)effort = loweffort = mediumeffort = higheffort = xhigheffort = maxeffort = nonebudget_tokens ≥ 1,024 ≤ 63,999

Interleaved thinking (reasoning between tool calls) is declared by 3 of 59 providers.

Your usage cost

Common scenarios
1IteraCompute$17.10
2Outer Labs$20.52
3Anthropic · Official$34.20
4Azure Cognitive Services$34.20
5Merge Gateway$34.20
6DevPass (LLM Gateway)$34.20

1 more channels offer $0 (GitLab Duo); free tiers usually have rate limits and no SLA, excluded from ranking.

Switch to IteraCompute to save $17.10/mo (50%)
Note: this is a gateway; verify availability and rate limits yourself.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.50$02026-10-082026-11-15
2026-11-15Input list $0.10Output list $0.50Min blended $0.10

Benchmark8 items

NameConditionsScoreMetricSource
GDPval-AAv2.11620EloSource ↗
AA-Briefcasev1.11578EloSource ↗
OSWorldvariant: offline subset · v2.172.4partial scoreSource ↗
Humanity's Last Examvariant: no tools45.9scoreSource ↗
Humanity's Last Examvariant: with tools57.4scoreSource ↗
Terminal-Benchv4.039.2scoreSource ↗
FrontierCodedataset: Main · v1.146.4scoreSource ↗
Chartographyvariant: no tools46.4scoreSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Artificial Analysis evaluations26 items

Humanity's Last Exam44.4%
HLE-Diamond29.6%
CritPt18.9%
CritPt v2 Hard43.5%
SciCode55.0%
FrontierCode41.8%
FrontierCode 2.034.9%
Terminal-Bench 4.032.8%
Terminal-Bench-Science20.0%
AutomationBench-AA35.4%
AA-AnalystAgent53.8%
AA-Briefcase1573
GDPval-AA1629
GDP.pdf20.8%
AA-LCR82.7%
MLCR-AA35.0%
AA-Omniscience Index12.3
AA-Omniscience Accuracy34.7%
AA-Omniscience Non-Hallucination65.8%
Harvey LAB-AA0.6%
HealthBench58.5%
HealthBench Professional58.4%
AA-3D998
Code2Video-AA1192
mka0.4096
vials-aa0.4907

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.

Related models

Claude Haiku Latestsame series—Claude Haiku 4.5 (latest)same series$1.00 / $5.00Claude Haiku 4.5same series$1.00 / $5.00Nemotron 3 Ultra 550B A55Bcheaper alternative$0 / $0Nemotron 3.5 Super VL 120B A12Bcheaper alternative$0 / $0reBeam 1.0 501B A23Bcheaper alternative$0 / $0Qwen3.7 Flashcheaper alternative$0.03 / $0.13
Data partly from models.dev (MIT) · Anthropic official docs ↗