AI Model Benchmark Directory

LMSpeed is an AI model directory for comparing API price, output speed, first-token latency, provider coverage, and benchmark data. Use it to narrow your model and provider choice, then test the endpoint that fits your workload.

Price, speed, latency, and availability can change. Treat this table as a current signal and verify your own endpoint before you deploy.

Newest Most tested Recently tested A–Z

Model	Context	Input	Output	Providers	Agents	Coding	Reasoning	Knowledge	Math	Multilingual	Multimodal	Instruction following	Throughput	Latency	Release date
Claude Sonnet 4.5Anthropic	Context1M	Input$3.00/M	Output$15.00/M	Providers +174	46.5±8.7	51.6±11.2E	45.5±8.9E	54.2±16.1P	42.5±12.3E	—	—	—	Throughput 41 t/s	Latency 4.57s	Release date2025-09-29	46.5 51.6 45.5 54.2 42.5 — — —
Claude Opus 4.6Anthropic	Context1M	Input$5.00/M	Output$25.00/M	Providers +229	54.8±5.7	58.7±8.0	53.4±8.7	53.7±13.7P	56.5±16.1P	54±17.3P	48.4±11.1E	44.7±16.0P	Throughput 45 t/s	Latency 5.30s	Release date2026-02-04	54.8 58.7 53.4 53.7 56.5 54 48.4 44.7
Claude Opus 4Anthropic	Context200K	Input$15.00/M	Output$75.00/M	Providers +91	—	51±13.9P	51.7±11.1E	—	54.4±11.8E	—	—	—	Throughput 25 t/s	Latency 12.05s	Release date2025-05-22	— 51 51.7 — 54.4 — — —
Claude Opus 4.5Anthropic	Context200K	Input$5.00/M	Output$25.00/M	Providers +166	48.4±5.3	55.2±8.1	60.6±8.1	57.6±11.6E	49.9±9.1E	51.7±12.2E	32.4±6.5	45.7±11.9E	Throughput 54 t/s	Latency 3.12s	Release date2025-11-24	48.4 55.2 60.6 57.6 49.9 51.7 32.4 45.7
Claude Opus 4.1Anthropic	Context200K	Input$15.00/M	Output$75.00/M	Providers +100	46.4±16.2P	46.9±16.1P	—	58.9±16.0P	—	—	—	—	Throughput 22 t/s	Latency 3.56s	Release date2025-08-05	46.4 46.9 — 58.9 — — — —
Claude 3.5 SonnetAnthropic	Context—	Input$3.00/M	Output$15.00/M	Providers +27	—	36.1±11.2E	44.3±11.1E	—	43±9.3E	—	—	—	Throughput 21 t/s	Latency 3.36s	Release date—	— 36.1 44.3 — 43 — — —
Claude 2.1Anthropic	Context—	Input—	Output—	Providers +8	—	33.7±13.9P	34.7±11.1E	—	38.5±11.8E	—	—	—	Throughput —	Latency —	Release date—	— 33.7 34.7 — 38.5 — — —
Claude Instant 1.2Anthropic	Context—	Input—	Output—	Providers +6	—	—	—	—	—	—	—	—	Throughput —	Latency —	Release date—	— — — — — — — —
Claude Haiku 4.5Anthropic	Context200K	Input$1.00/M	Output$5.00/M	Providers +198	42.5±16.2P	46.4±11.2E	47.5±11.1E	—	45.7±16.1P	—	—	—	Throughput 102 t/s	Latency 2.77s	Release date2025-10-15	42.5 46.4 47.5 — 45.7 — — —
Claude Sonnet 4Anthropic	Context1M	Input$3.00/M	Output$15.00/M	Providers +121	44.4±9.3E	46.3±11.2E	47.4±8.7	49.2±16.0P	52.9±11.8E	—	—	45±16.0P	Throughput 34 t/s	Latency 4.93s	Release date2025-05-22	44.4 46.3 47.4 49.2 52.9 — — 45
Claude 3.5 HaikuAnthropic	Context200K	Input—	Output—	Providers +28	—	41.4±13.9P	39.3±11.1E	—	42.5±11.8E	—	—	—	Throughput 42 t/s	Latency 2.46s	Release date2024-11-04	— 41.4 39.3 — 42.5 — — —
Claude Sonnet 4.6Anthropic	Context1M	Input$3.00/M	Output$15.00/M	Providers +223	52.5±6.1	55.7±8.5	49.8±8.7	70.6±16.3P	53.2±16.1P	—	45.7±16.1P	43.7±16.0P	Throughput 66 t/s	Latency 4.16s	Release date2026-02-17	52.5 55.7 49.8 70.6 53.2 — 45.7 43.7
Claude 3.7 SonnetAnthropic	Context—	Input$3.00/M	Output$15.00/M	Providers +36	—	47.5±13.9P	47.7±11.1E	—	48.7±11.8E	—	—	—	Throughput 45 t/s	Latency 7.65s	Release date—	— 47.5 47.7 — 48.7 — — —
Claude 3 SonnetAnthropic	Context—	Input—	Output—	Providers +8	—	36.3±13.9P	37.7±11.1E	—	39.6±11.8E	—	—	—	Throughput —	Latency —	Release date—	— 36.3 37.7 — 39.6 — — —
Claude 3 OpusAnthropic	Context—	Input$15.00/M	Output$75.00/M	Providers +11	—	38.5±13.9P	41.5±11.1E	54.5±16.0P	41.5±11.8E	—	—	—	Throughput —	Latency —	Release date—	— 38.5 41.5 54.5 41.5 — — —
Claude Opus 4.6 MaxAnthropic	Context—	Input—	Output—	Providers +21	—	—	—	—	—	—	—	—	Throughput —	Latency —	Release date—	— — — — — — — —
Claude 3 HaikuAnthropic	Context200K	Input$0.250/M	Output$1.25/M	Providers +15	38±16.0P	32.9±13.9P	36±10.8E	39.2±16.0P	36.4±11.8E	—	—	42.2±16.0P	Throughput —	Latency —	Release date2024-03-13	38 32.9 36 39.2 36.4 — — 42.2
Claude 2.0Anthropic	Context—	Input—	Output—	Providers +8	—	33.9±13.9P	34.8±11.8E	—	33.3±16.0P	—	—	—	Throughput —	Latency —	Release date—	— 33.9 34.8 — 33.3 — — —
Claude Opus 4.7Anthropic	Context1M	Input$5.00/M	Output$25.00/M	Providers +198	54.3±6.9	59.9±11.2E	55.8±10.8E	54.3±14.0P	56.8±16.1P	—	—	44.5±16.0P	Throughput 47 t/s	Latency 4.89s	Release date2026-05-12	54.3 59.9 55.8 54.3 56.8 — — 44.5
Claude Opus 4.7 MaxAnthropic	Context—	Input—	Output—	Providers +20	57±7.7	60.5±6.9	56.4±8.7	61.3±14.0P	48.3±17.5P	—	60.7±16.1P	48.8±16.0P	Throughput —	Latency —	Release date—	57 60.5 56.4 61.3 48.3 — 60.7 48.8
Claude Opus 4.8Anthropic	Context1M	Input$5.00/M	Output$25.00/M	Providers +156	62.4±5.5	64.7±6.6	56.1±8.7	63.1±14.0P	58±16.1P	55±17.1P	59.6±12.0E	49.9±16.0P	Throughput 193 t/s	Latency 2.96s	Release date2026-05-27	62.4 64.7 56.1 63.1 58 55 59.6 49.9
Claude Fable 5Anthropic	Context1M	Input$10.00/M	Output$50.00/M	Providers +122	64.2±8.2	66.9±6.9	60.5±10.8E	59.6±14.0P	—	—	46.7±17.3P	50.3±16.0P	Throughput 61 t/s	Latency 3.92s	Release date2026-06-09	64.2 66.9 60.5 59.6 — — 46.7 50.3
Claude Sonnet 5Anthropic	Context1M	Input$2.00/M	Output$10.00/M	Providers +112	63.7±12.2P	59.2±6.6	56.2±10.8E	60.3±14.0P	—	—	56.6±16.1P	—	Throughput 158 t/s	Latency 6.92s	Release date2026-06-30	63.7 59.2 56.2 60.3 — — 56.6 —
Claude Opus 4.8 (Fast)Anthropic	Context1M	Input—	Output—	Providers	—	—	—	—	—	—	—	—	Throughput —	Latency —	Release date2026-05-27	— — — — — — — —
Claude Opus 4.7 (Fast)Anthropic	Context1M	Input—	Output—	Providers	—	—	—	—	—	—	—	—	Throughput —	Latency —	Release date2026-05-12	— — — — — — — —
Claude Opus 5Anthropic	Context1M	Input$5.00/M	Output$25.00/M	Providers +84	67.4±7.7	67.5±6.6	59.9±8.7	67.7±14.0P	—	64.3±17.1P	56±17.3P	—	Throughput 284 t/s	Latency 5.44s	Release date2026-07-24	67.4 67.5 59.9 67.7 — 64.3 56 —
Claude Fable 5.1Anthropic	Context1M	Input$10.00/M	Output$50.00/M	Providers	64.4±8.6	67.9±6.9	61.3±8.7	68.7±14.0P	—	—	—	—	Throughput —	Latency —	Release date2026-09-01	64.4 67.9 61.3 68.7 — — — —

How to read category scores

Agents, Coding, Reasoning, and the other capability columns are 0–100 observed-capability estimates relative to the eligible model population in a dated Category Score V3 run.

They are not success rates, IQ scores, or an average across all eight categories. Read them with the 80% interval, measured dimensions, benchmark families, and evidence shown on each model page.

Read the complete scoring methodology

What this directory shows

Each row brings together the information you need to compare an LLM API. Some fields are blank when LMSpeed has no current data for that model or provider.

API price: Compare input and output price per million tokens when it is available.
Speed and latency: Use throughput and first-token latency to compare response behavior.
Provider coverage: Open a model to review its listed providers and their current details.
Capability data: Use the capability columns when current model scores are available.

How to use the model directory

Start with the decision that matters most for your workload, then compare the current rows before you test an endpoint.

Find a model.Search by model name, slug, or description.
Sort the key metric.Sort by price, throughput, latency, provider count, or capability data.
Compare providers.Open a model page to review the provider options and current data.
Test your endpoint.Run a speed test before you use an endpoint in production.

Model pricing leaderboard Model speed leaderboard Latency leaderboard Provider directory LLM API speed test

Frequently Asked Questions

How does LMSpeed benchmark AI models?

LMSpeed runs standardized five-round API speed tests on each model, measuring output throughput (tokens per second), first-token latency, and total response time across multiple providers.

Which AI model has the lowest API latency?

Latency varies by provider and model. Use the LMSpeed model directory to sort by latency and find the model with the fastest first-token response time. Check the latency leaderboard for monthly rankings.

How to compare LLM API pricing across providers?

LMSpeed lists input and output token prices per million for each model across available providers. Sort by price to find a lower-cost option, or filter by provider to compare rates side by side.

What is an AI model directory?

An AI model directory is a searchable list of models and their comparison data. On LMSpeed, it brings together API price, throughput, first-token latency, provider coverage, and capability data when available.

How do I choose a model and provider?

Start with the metric that matters most for your workload. Then open a model page to compare provider data. Run your own speed test before you use an endpoint in production.

Can model price and speed change?

Yes. Price, availability, throughput, and latency can change by model, provider, and time. Use current page values as a comparison signal and confirm with your own endpoint test.