Top for agents
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.
| Release date | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Context128K | Input— | Output— | Providers |
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.
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.
Start with the decision that matters most for your workload, then compare the current rows before you test an endpoint.
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.
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.
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.
— |
— |
— |
— |
— |
— |
— |
— |
| Throughput — | Latency — |
| Release date2026-04-27 |
|
| Laguna M 1Poolside | Context262.1K | Input— | Output— | Providers +8 | 43.9±16.0P | 40.7±9.0 | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-04-28 |
|
| Laguna Xs 2Poolside | Context262.1K | Input— | Output— | Providers +8 | 39.8±16.0P | 35.3±9.0 | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-04-28 |
|
| Nemotron 3 Nano OmniNVIDIA | Context256K | Input$0.075/M | Output$0.300/M | Providers +9 | — | 44.1±16.0P | 41.6±13.9P | — | — | — | — | — | Throughput — | Latency — | Release date2026-04-28 |
|
| Mistral Medium 3.5Mistral | Context262.1K | Input$1.50/M | Output$7.50/M | Providers +10 | 44.6±6.3 | 48.4±9.4E | 50±10.8E | 43.3±16.0P | — | — | — | 52.4±16.0P | Throughput 47 t/s | Latency 10.40s | Release date2026-04-30 |
|
| Granite 4.1 8BIBM | Context131.1K | Input$0.050/M | Output$0.100/M | Providers | — | 38.8±16.0P | 39.6±13.9P | — | — | — | — | — | Throughput — | Latency — | Release date2026-04-30 |
|
| Grok 4.3xAI | Context1M | Input$1.25/M | Output$2.50/M | Providers +49 | 49.6±6.6 | 56.1±16.0P | 56.3±10.8E | 52.6±14.0P | — | — | 49.5±16.1P | 57.6±16.0P | Throughput 49 t/s | Latency 6.49s | Release date2026-04-30 |
|
| GPT-4o Mini TranscribeOpenAI | Context128K | Input— | Output— | Providers +35 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-05-01 |
|
| Gemini 3.1 Flash LiteGoogle | Context1.0M | Input— | Output— | Providers +108 | 42.8±6.9 | 40.9±11.8E | 52.8±10.8E | 48.7±16.0P | — | — | 39.5±16.2P | 55.7±16.0P | Throughput 191 t/s | Latency 6.58s | Release date2026-05-07 |
|
| Ring 2 6 1tinclusionAI | Context262.1K | Input$0.300/M | Output$2.50/M | Providers +2 | — | 54.7±16.0P | 55.6±13.9P | — | — | — | — | — | Throughput — | Latency — | Release date2026-05-08 |
|
| Claude Opus 4.7Anthropic | Context1M | Input$5.00/M | Output$25.00/M | Providers +181 | 55±6.9 | 60.3±11.2E | 56.6±10.8E | 54±14.0P | 56.8±16.1P | — | — | 44.5±16.0P | Throughput 47 t/s | Latency 4.89s | Release date2026-05-12 |
|
| Gemini 3.5 FlashGoogle | Context1.0M | Input$1.50/M | Output$9.00/M | Providers +81 | 59.5±5.2 | 54.7±8.9 | 54.6±8.5 | 56.1±14.0P | 55.1±16.1P | — | 58.5±11.9E | 55.3±16.0P | Throughput — | Latency — | Release date2026-05-19 |
|
| Gemini Embedding 2Google | Context8.2K | Input— | Output— | Providers +34 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-05-20 |
|
| Grok Build 0 1xAI | Context256K | Input$1.00/M | Output$2.00/M | Providers +28 | 51.7±11.8E | 59.8±16.0P | 57.1±10.8E | 54.7±16.0P | — | — | — | — | Throughput 131 t/s | Latency 12.87s | Release date2026-05-20 |
|
| Qwen3.7 MaxQwen | Context1M | Input$2.50/M | Output$7.50/M | Providers +28 | 58±5.5 | 59.5±6.0 | 60.1±8.7 | 55.8±12.2E | 64.5±12.4E | 59.9±11.7E | — | 57.1±11.9E | Throughput 78 t/s | Latency 14.81s | Release date2026-05-21 |
|
| Claude Opus 4.8Anthropic | Context1M | Input$5.00/M | Output$25.00/M | Providers +132 | 63.7±5.3 | 66±6.6 | 59.9±10.8E | 64.4±14.0P | 58±16.1P | 61.2±17.5P | 62.7±12.0E | 50.2±16.0P | Throughput 232 t/s | Latency 2.21s | Release date2026-05-27 |
|
| Step 3.7 FlashStepFun | Context256K | Input$0.200/M | Output$1.15/M | Providers +30 | 54.6±5.7 | 49.7±9.3E | 53.3±10.8E | 43±16.0P | — | — | 59.5±14.6P | 51.9±16.0P | Throughput 119 t/s | Latency 15.57s | Release date2026-05-28 |
|
| MiniMax M3MiniMax | Context1.0M | Input$0.300/M | Output$1.20/M | Providers +59 | 56.9±7.3 | 53±6.6 | 59.5±10.8E | 52.9±14.0P | — | — | 47.1±12.0E | 58.4±16.0P | Throughput 62 t/s | Latency 2.10s | Release date2026-05-31 |
|
| Qwen3.7 PlusQwen | Context1M | Input$0.400/M | Output$1.60/M | Providers +31 | 54.3±5.7 | 53.8±6.0 | 58.1±8.7 | 48.6±12.2E | 57.2±12.4E | 51.9±11.7E | 57.5±6.3 | 57.1±11.9E | Throughput — | Latency — | Release date2026-06-03 |
|
| Nemotron 3 UltraNVIDIA | Context1M | Input$0.675/M | Output$2.67/M | Providers +9 | — | 53±16.0P | 57.7±13.9P | — | — | — | — | — | Throughput — | Latency — | Release date2026-06-04 |
|
| Nex n2 ProNex AGI | Context262.1K | Input$0.500/M | Output$2.50/M | Providers +3 | — | 54.3±16.0P | 59.9±13.9P | — | — | — | — | — | Throughput 83 t/s | Latency 1.16s | Release date2026-06-08 |
|
| Claude Fable 5Anthropic | Context1M | Input$10.00/M | Output$50.00/M | Providers +101 | 66.7±8.2 | 67.9±6.9 | 61.2±10.8E | 61.1±14.0P | — | — | 49±18.4P | 50.6±16.0P | Throughput 62 t/s | Latency 3.91s | Release date2026-06-09 |
|
| Kimi K2.7Moonshot AI | Context262.1K | Input— | Output— | Providers +30 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-06-12 |
|
| Kimi K2.7 CodeMoonshotAI | Context262.1K | Input$0.950/M | Output$4.00/M | Providers +28 | 54.5±11.2E | 58±16.0P | 57.4±10.8E | 50±16.0P | — | — | — | 50.5±16.0P | Throughput 46 t/s | Latency 3.67s | Release date2026-06-12 |
|
| GLM-5.2Z.ai | Context1.0M | Input$1.40/M | Output$4.40/M | Providers +114 | 61.4±7.2 | 58±8.9 | 55±10.8E | 58.8±14.0P | 70.4±11.5E | — | — | 54.1±16.0P | Throughput 69 t/s | Latency 6.95s | Release date2026-06-16 |
|
| North Mini CodeCohere | Context256K | Input— | Output— | Providers +6 | — | 51.8±16.0P | 50±13.9P | — | — | — | — | — | Throughput — | Latency — | Release date2026-06-17 |
|
| Gemini 3 Pro ImageGoogle | Context65.5K | Input— | Output— | Providers +103 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-06-18 |
|
| Gemini 3.1 Flash ImageGoogle | Context131.1K | Input— | Output— | Providers +119 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-06-18 |
|
| Claude Sonnet 5Anthropic | Context1M | Input$2.00/M | Output$10.00/M | Providers +81 | 60.7±8.2 | 59.8±6.6 | 56.8±10.8E | 62±14.0P | — | — | 62.3±16.2P | — | Throughput — | Latency — | Release date2026-06-30 |
|
| Hy3Tencent | Context262.1K | Input— | Output— | Providers +10 | 53.4±16.0P | 58.1±16.0P | 56.8±10.8E | 47.9±16.0P | — | — | — | — | Throughput — | Latency — | Release date2026-07-06 |
|
| Grok 4.5xAI | Context500K | Input$2.00/M | Output$6.00/M | Providers +51 | 62.9±8.4 | 59.4±6.9 | 59.4±10.8E | 58.3±14.0P | — | — | 51.1±18.4P | — | Throughput — | Latency — | Release date2026-07-08 |
|
| GPT-5.6 SolOpenAI | Context1.1M | Input$5.00/M | Output$30.00/M | Providers +93 | 68.4±7.4 | 62.9±9.3E | 62±10.8E | 57.2±16.7P | 67±16.1P | — | 60.9±16.1P | 53.9±16.0P | Throughput — | Latency — | Release date2026-07-09 |
|
| GPT-5.6 TerraOpenAI | Context1.1M | Input$2.50/M | Output$15.00/M | Providers +91 | 63.8±7.5 | 59.8±9.3E | 60.7±10.8E | 57±16.7P | 64.5±16.1P | — | 55.2±16.1P | 53.3±16.0P | Throughput — | Latency — | Release date2026-07-09 |
|
| GPT-5.6 LunaOpenAI | Context1.1M | Input$1.00/M | Output$6.00/M | Providers +84 | 61±7.6 | 57.3±9.3E | 58.9±10.8E | 56.5±16.7P | 62.4±16.1P | — | 50.1±16.1P | — | Throughput — | Latency — | Release date2026-07-09 |
|
| DeepSeek V3 | Context— | Input$0.360/M | Output$0.890/M | Providers +147 | 35.5±11.8E | 37.9±11.2E | 44.8±8.7 | 42±16.0P | 49.5±9.3E | — | — | 39.7±11.9E | Throughput 49 t/s | Latency 3.39s | Release date— |
|
| Qwen3 VL | Context— | Input— | Output— | Providers +17 | — | — | — | — | — | — | — | — | Throughput 206 t/s | Latency 6.63s | Release date— |
|
| Devstral 2 | Context— | Input— | Output— | Providers +3 | — | 46.8±13.9P | 45.6±11.1E | — | 43.3±16.1P | — | — | — | Throughput 71 t/s | Latency 1.34s | Release date— |
|
| Ministral 3 | Context— | Input$0.200/M | Output$0.200/M | Providers +1 | — | 32.1±13.9P | 37±11.1E | — | 40.6±16.1P | — | — | — | Throughput 208 t/s | Latency 2.84s | Release date— |
|
| Mistral Large 3 | Context— | Input$0.500/M | Output$1.50/M | Providers +5 | 40.1±11.8E | 48.5±13.9P | 45.3±8.7 | 42.5±16.0P | 43.5±16.1P | — | — | 42.2±16.0P | Throughput 29 t/s | Latency 3.00s | Release date— |
|
| Qwen3 Embedding | Context— | Input— | Output— | Providers +37 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| Qwen3 Next | Context— | Input— | Output— | Providers +7 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek Coder | Context— | Input— | Output— | Providers +8 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| Qwen Image | Context— | Input— | Output— | Providers +33 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| Intern-S1 | Context— | Input— | Output— | Providers +3 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| MAI-DS-R1 | Context— | Input— | Output— | Providers +9 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek VL2 | Context— | Input— | Output— | Providers +8 | — | — | — | — | — | — | — | — | Throughput 126 t/s | Latency 0.75s | Release date— |
|
| GLM-4 | Context— | Input— | Output— | Providers +51 | — | — | — | — | — | — | — | — | Throughput 63 t/s | Latency 0.72s | Release date— |
|
| MiniMax Hailuo 2.3 | Context— | Input— | Output— | Providers +14 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| O1 MiniOpenAI | Context— | Input— | Output— | Providers +57 | — | 48.1±13.9P | 45.6±11.1E | — | 54.9±11.8E | — | — | — | Throughput 32 t/s | Latency 13.84s | Release date— |
|
| Gemini 3 ProGoogle | Context— | Input$2.00/M | Output$12.00/M | Providers +131 | 49.5±9.3E | 55.7±11.2E | 54.4±6.5 | 56.3±14.0P | 55.7±16.1P | 54±17.3P | 51.8±6.5 | 53±16.0P | Throughput 73 t/s | Latency 10.59s | Release date— |
|
| Gemini 1.5 ProGoogle | Context— | Input— | Output— | Providers +20 | — | 40.9±13.9P | 40.2±11.1E | 55.4±16.0P | 43.5±11.8E | — | — | — | Throughput 18 t/s | Latency 2.58s | Release date— |
|
| Devstral SmallMistral AI | Context— | Input$0.100/M | Output$0.300/M | Providers +6 | — | 39.5±13.9P | 40.1±11.1E | — | 43.2±11.8E | — | — | — | Throughput — | Latency — | Release date— |
|
| Qwen3 Omni Flash | Context— | Input— | Output— | Providers +35 | — | — | — | — | — | — | — | — | Throughput 62 t/s | Latency 5.13s | Release date— |
|
| DeepSeek ReasonerDeepSeek | Context— | Input— | Output— | Providers +83 | — | — | — | — | — | — | — | — | Throughput 28 t/s | Latency 6.40s | Release date— |
|
| Claude 3.5 SonnetAnthropic | Context— | Input$3.00/M | Output$15.00/M | Providers +24 | — | 36.8±11.2E | 44.9±11.1E | — | 42.9±9.3E | — | — | — | Throughput 21 t/s | Latency 3.36s | Release date— |
|
| GPT-4.5OpenAI | Context— | Input— | Output— | Providers +38 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| Qwen3 S2S Flash Realtime | Context— | Input— | Output— | Providers +12 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| GPT-SoVITS | Context— | Input— | Output— | Providers +1 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| Gemini 1.5 Pro 001Google | Context— | Input— | Output— | Providers +1 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| GPT-4 VisionOpenAI | Context— | Input— | Output— | Providers +31 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
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.
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.
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.
