Anthropic
·Released on Nov 24, 2025Claude Opus 4.5 API Benchmarks, Pricing & Provider Data
Compare Claude Opus 4.5 with another model
Choose a model to open its comparison page.
LLM
Claude Opus 4.5 benchmark, API pricing, and provider data cover 573 API providers, with prices starting at $0.0068/request. Claude Opus 4.5 free API options are available from 6 providers. The page also shows measured API speed and first-token latency.
Anthropic Claude Opus 4.5 is the flagship Claude model for the hardest reasoning, research, and multi-step agent tasks where accuracy matters more than speed.
Quality
#56of 101
55.0
LMSpeed score
Speed
54char/s
3.12 s
Cost
#165of 182
$0.0068/ 1M · 8:1 in:out
$0.0011 in · $0.0058 out
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 8 / 8
- Methodology
- V3.0
#1Reasoning60.6RatedGlobal rank #43/4 Measured dimensions
80% interval: 52.5–68.8
abstract logicPrior only
scientific causal53.3
critpt · critpt / gpqa · gpqa / hle · hle
multistep constraints61.7
mmlu_pro · mmlu_pro
evidence verification66.9
ai_needle · ai_needle / lcr · lcr / longbench · long_bench_v2
#2Knowledge57.6Estimated2/4 Measured dimensions
80% interval: 45.9–69.2
broad knowledge65.7
c_eval · c_eval / mmlu_redux · mmlu_redux
professional knowledge49.4
supergpqa · super_gpqa
factualityPrior only
retrieval open bookPrior only
#3Coding55.2RatedGlobal rank #163/4 Measured dimensions
80% interval: 47.1–63.3
code generation59.1
livecodebench · live_code_bench_v6 / livecodebench · livecodebench / scicode · scicode
repository engineering48.7
nl2repo · nl2_repo / swe_pro · swe_pro
debugging testing57.9
swe_multilingual · benchlm_coding_swe_multilingual / swe_verified · swe_verified
tooling qualityPrior only
#4Multilingual51.7Estimated2/4 Measured dimensions
80% interval: 39.5–63.9
cross language understanding45.3
nova · nova63
multilingual generationPrior only
reasoning transfer58.1
mmlu_prox · mmlu_pro_x
low resource robustnessPrior only
#5Math49.9Estimated3/4 Measured dimensions
80% interval: 40.8–58.9
foundational mathPrior only
competition math46.5
aime · aime2026 / hmmt · hmmt_feb2025 / hmmt · hmmt_feb2026 / hmmt · hmmt_nov2025
proof frontier50.3
frontiermath · frontier_math_v2_tier4 / frontiermath · frontier_math_v2_tiers13
applied tool math52.8
mmanswer · mm_answer_bench
#6Agents48.4RatedGlobal rank #394/4 Measured dimensions
80% interval: 43.1–53.8
planning57.9
deep_planning · gert_labs
tool use42.6
mcp_atlas · mcp_atlas / mcp_tasks · mcp_tasks / tau · tau2_bench / tau · tau3_bench / toolathlon · toolathlon
environment execution48.1
osworld · os_world_verified / terminalbench · benchlm_agentic_terminal_bench2 / vita_bench · vita_bench / wide_research · wide_research
recovery reliability45.1
claw_eval · claw_eval / cyber_gym · cyber_gym / jobbench · job_bench
#7Instruction following45.7Estimated2/4 Measured dimensions
80% interval: 33.8–57.6
constraint following47.1
ifeval · ifeval
structured outputPrior only
novel instruction generalization44.3
ifbench · aa_if_bench
long multiturn instructionPrior only
#8Multimodal32.4RatedGlobal rank #104/4 Measured dimensions
80% interval: 26.0–38.9
perception ocr29.4
v_star · v_star
document spatial39.4
charxiv · charxiv
visual reasoning26.3
mathvision · math_vision / mmmu_pro · mmmu_pro
video action34.5
screenspot · screen_spot_pro / video_mmmu · video_mmmu
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
INPUT
200Ktokens
≈ 240 pages of text
OUTPUT
64Ktokens
8K128K1M4M
200K
Features
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Sep 2, 2026Overall
undefined metric} other undefined metrics}}
Overall score55.0#56 / 101
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Pricing
undefined metric} other undefined metrics}}
Input price$5.00/M#165 / 182Output price$25.00/M#166 / 182
Pricing
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score48.4#3980% interval43.1–53.84/4 Measured dimensionsAgentic score48.1#53 / 70Terminal-Bench 2.059.3#33 / 53
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Planning & decomposition57.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_planning · gert_labs · z 1.07 · q 1.00
Tool use42.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z -2.31 · q 1.00
mcp_tasks · mcp_tasks · z 0.00 · q 0.63
tau · tau2_bench · z 0.01 · q 1.00
tau · tau3_bench · z 0.00 · q 1.00
toolathlon · toolathlon · z -0.46 · q 1.00
Environment & long-horizon execution48.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
osworld · os_world_verified · z -0.59 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z -0.28 · q 1.00
vita_bench · vita_bench · z -0.64 · q 1.00
wide_research · wide_research · z 0.89 · q 1.00
Recovery & completion reliability45.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
claw_eval · claw_eval · z 0.00 · q 1.00
cyber_gym · cyber_gym · z -2.03 · q 1.00
jobbench · job_bench · z 0.00 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Score55.2#1680% interval47.1–63.33/4 Measured dimensionsLiveCodeBench73.8%#24 / 115SciCode47.0%#36 / 207
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Code generation59.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
livecodebench · live_code_bench_v6 · z -0.38 · q 0.75
livecodebench · livecodebench · z 1.29 · q 1.00
scicode · scicode · z 1.02 · q 1.00
Repository engineering48.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
nl2repo · nl2_repo · z 0.08 · q 1.00
swe_pro · swe_pro · z 0.03 · q 1.00
Debugging & testing57.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_multilingual · benchlm_coding_swe_multilingual · z 0.38 · q 1.00
swe_verified · swe_verified · z 0.87 · q 1.00
Tool-assisted development & qualityPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
LiveCodeBenchV3 evidence
LMSpeed rank#24
Score73.8%
Sourceartificialanalysis.ai
UpdatedSep 2, 2026
Confidence4
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Score60.6#480% interval52.5–68.83/4 Measured dimensionsMMLU-Pro88.9%#4 / 129GPQA81.0%#92 / 214
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning53.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z -0.72 · q 1.00
gpqa · gpqa · z 0.62 · q 1.00
hle · hle · z 0.65 · q 1.00
Multi-step constraints61.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmlu_pro · mmlu_pro · z 1.35 · q 1.00
Evidence integration & verification66.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ai_needle · ai_needle · z 2.61 · q 0.50
lcr · lcr · z -0.19 · q 1.00
longbench · long_bench_v2 · z 1.78 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Estimated
Score57.680% interval45.9–69.22/4 Measured dimensionsKnowledge score58.0#51 / 70SuperGPQA70.6#7 / 16
Knowledge
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Broad knowledge65.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
c_eval · c_eval · z 0.00 · q 0.63
mmlu_redux · mmlu_redux · z 3.00 · q 0.75
Professional knowledge49.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
supergpqa · super_gpqa · z 0.18 · q 1.00
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence Index
LMSpeed rank#60
Score35.6
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
AA-Omniscience Hallucination Rate
LMSpeed rank#48
Score76.2
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Score49.980% interval40.8–58.93/4 Measured dimensionsMath score58.4#29 / 62AIME2695.1#9 / 14
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Foundational mathPrior only
Competition math46.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aime · aime2026 · z -0.18 · q 1.00
hmmt · hmmt_feb2025 · z -0.80 · q 0.88
hmmt · hmmt_feb2026 · z -0.30 · q 1.00
hmmt · hmmt_nov2025 · z 0.00 · q 1.00
Advanced proofs50.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
frontiermath · frontier_math_v2_tier4 · z -0.24 · q 1.00
frontiermath · frontier_math_v2_tiers13 · z -0.11 · q 1.00
Applied & tool-assisted math52.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmanswer · mm_answer_bench · z 0.42 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
FrontierMath v2 (Tiers 1-3)V3 evidence
LMSpeed rank#29
Score20.7
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
FrontierMath v2 (Tier 4)V3 evidence
LMSpeed rank#23
Score4.2
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
Multilingual
V3.0undefined metric} other undefined metrics}} · Estimated
Score51.780% interval39.5–63.92/4 Measured dimensionsMultilingual score82.9#2 / 11MMLU-ProX85.7#2 / 11
Multilingual
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Cross-language understanding45.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
nova · nova63 · z -0.63 · q 0.88
Multilingual generationPrior only
Reasoning transfer58.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmlu_prox · mmlu_pro_x · z 1.28 · q 1.00
Low-resource robustnessPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Score32.4#1080% interval26.0–38.94/4 Measured dimensionsMultimodal Grounded score13.5#43 / 44MMMU-Pro70.6#29 / 31
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Perception & OCR29.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
v_star · v_star · z -3.00 · q 1.00
Document & spatial understanding39.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
charxiv · charxiv · z -1.29 · q 1.00
Visual reasoning26.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mathvision · math_vision · z -3.00 · q 1.00
mmmu_pro · mmmu_pro · z -1.72 · q 1.00
Video & grounded action34.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
screenspot · screen_spot_pro · z -2.46 · q 1.00
video_mmmu · video_mmmu · z -0.17 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Estimated
Score45.780% interval33.8–57.62/4 Measured dimensionsInstruction Following score56.4#23 / 25IFEval90.9#12 / 16
Instruction following
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Constraint following47.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifeval · ifeval · z -0.59 · q 1.00
Structured outputPrior only
Novel-instruction generalization44.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z -1.04 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
OpenRouter endpoints
6 endpointsThird-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.
| Provider endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
Azure azure/global | $5/M | $25/M | 100% | — | — | undefined tokens / undefined tokens |
Amazon Bedrock amazon-bedrock/eu-west-1 | $5.50/M | $27.50/M | 100% | — | — | undefined tokens / undefined tokens |
Google google-vertex/global | $5/M | $25/M | 100.0% | — | — | undefined tokens / undefined tokens |
Amazon Bedrock amazon-bedrock | $5/M | $25/M | 99.9% | — | — | undefined tokens / undefined tokens |
Anthropic anthropic | $5/M | $25/M | 99.8% | — | — | undefined tokens / undefined tokens |
Claude Platform on AWS claude-on-aws | $5/M | $25/M | 99.7% | — | — | undefined tokens / undefined tokens |
Pricing Comparison
Compare Claude Opus 4.5 API pricing across 567 providers. Prices range from $0.0068/request to $8000.00/M. 钠 API offers the lowest rate at $0.0068/request. 6 providers offer free API credits or a free tier.
| Provider | Health | Model Variant | Group | Input ($/M) | Output ($/M) | Speed (t/s) | First token | Audit |
|---|---|---|---|---|---|---|---|---|
L1 100% | claude-opus-4-5-20251101 | default | -86%$0.685/M Cache read$0.068/M | -86%$3.42/M | 56.5 t/s | 2.60 s | — | |
L1 100% | claude-opus-4-5-20251101 | aws-q | -96%$0.214/M Cache read$0.021/M | -96%$1.07/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | -93%$0.342/M Cache read$0.034/M | -93%$1.71/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | claude_kiro | -65%$1.75/M | -65%$8.75/M | — | — | — | |
L1 99% | claude-opus-4-5-20251101 | aws | -99%$0.059/M Cache read$0.0059/MCache write$0.073/MCache write 1h$0.117/M | -99%$0.294/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | -86%$0.685/M Cache read$0.068/MCache write$0.856/MCache write 1h$1.37/M | -86%$3.42/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | Kiro | -97%$0.137/M | -97%$0.685/M | — | — | — | |
L1 100% | claude-opus-4-5 | default | -86%$0.685/M | -86%$3.42/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | -86%$0.685/M | -86%$3.42/M | — | — | — | |
L1 100% | claude-opus-4.5 | default | -59%$2.05/M | -59%$10.27/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | $5.00/M Cache read$0.500/MCache write$6.25/MCache write 1h$10.00/M | $25.00/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | -23%$3.84/M | -23%$19.18/M | — | — | — | |
兔子API Free | L1 100% | claude-opus-4-5-20251101 | default | Free | Free | — | — | — |
L1 100% | claude-opus-4-5 | Claude-Code | Free | Free | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | claude-huaqiangbei | -95%$0.250/M Cache read$0.025/M | -95%$1.25/M | — | — | — | |
L1 100% | claude-opus-4-5 | claude-kiro-sale | -90%$0.500/M | -90%$2.50/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101-c | default | $0.428/request | - | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | default | $5.00/M Cache read$0.500/MCache write$6.25/MCache write 1h$10.00/M | $25.00/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | cc-sale | -89%$0.548/M Cache read$0.055/MCache write$0.685/MCache write 1h$1.10/M | -89%$2.74/M | — | — | — | |
L1 100% | claude-opus-4-5-20251101 | claude-MAX1-企业版 | -82%$0.923/M Cache read$0.092/MCache write$1.15/MCache write 1h$1.85/M | -82%$4.61/M | — | — | — |
Showing 20 model IDs of 120.
Alternatives & Similar Models
Claude Opus 4.6
claude-opus-4-6
Anthropic Claude Opus 4.6 is the most capable Claude Opus tier, optimized for complex analysis, long-horizon coding, and high-stakes enterprise reasoning workloads.
206 shared providers
Claude Sonnet 4.6
claude-sonnet-4-6
Anthropic Claude Sonnet 4.6 extends the Sonnet line with improved tool use, coding reliability, and long-context performance for everyday production workloads.
203 shared providers
Claude Sonnet 4.5
claude-sonnet-4-5
Anthropic Claude Sonnet 4.5 balances strong reasoning, coding, and agent capabilities with lower latency and cost than Opus, making it a practical default for production assistants.
203 shared providers
Claude Haiku 4.5
claude-haiku-4-5
Anthropic Claude Haiku 4.5 delivers fast, low-cost responses while retaining solid instruction following for chat, classification, and lightweight coding.
202 shared providers
GPT-5.4
gpt-5-4
OpenAI GPT-5.4 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
200 shared providers
GPT-5.3 Codex
gpt-5-3-codex
OpenAI GPT-5.3 Codex is a code-specialized variant in the GPT-5 series, optimized for code generation, debugging, and software development tasks.
185 shared providers
Frequently Asked Questions
- What benchmark data does Claude Opus 4.5 include?
- LMSpeed shows Claude Opus 4.5 benchmark context, API price, output speed, first-token latency, and provider data across 573 providers when those signals are available.
- What is the Claude Opus 4.5 API price?
- Claude Opus 4.5 has pricing from 573 providers, ranging from $0.0068/request to $8000.00/M. 钠 API has the lowest listed price.
- What does the Claude Opus 4.5 API pricing table include?
- The Claude Opus 4.5 API pricing table compares 573 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
- Which provider has the cheapest Claude Opus 4.5 API pricing?
- 钠 API currently has the lowest listed Claude Opus 4.5 price at $0.0068/request across 573 providers.
- Can I compare Claude Opus 4.5 API price and speed together?
- Yes. LMSpeed shows Claude Opus 4.5 API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
- Is Claude Opus 4.5 API free?
- Yes, Claude Opus 4.5 free API options are available through 6 providerundefined other undefined} on LMSpeed, including 兔子API, 兔子API, 兔子API, 兔子API, 兔子API. These providers offer free API credits or a free tier with no per-token charges.
- Where can I get Claude Opus 4.5 free API access?
- LMSpeed currently lists 6 free API providerundefined other undefined} for Claude Opus 4.5: 兔子API, 兔子API, 兔子API, 兔子API, 兔子API. Check each provider row before using it because free tier limits can change.