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 | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Context1.0M | Input$0.435/M | Output$0.870/M | 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.
52.6±5.2 |
49.4±6.3 |
53.8±8.4 |
44±14.0P |
35.9±12.4E |
— |
— |
53.3±16.0P |
| Throughput 38 t/s | Latency 8.33s |
| Release date2026-04-24 |
|
| DeepSeek V4 FlashDeepSeek | Context1.0M | Input$0.140/M | Output$0.280/M | Providers +190 | 48.5±5.3 | 44.5±6.6 | 51.6±8.4 | 39.1±14.0P | 36.6±12.4E | — | — | 54.2±16.0P | Throughput 66 t/s | Latency 7.28s | Release date2026-04-24 |
|
| Aion 2.0AionLabs | Context131.1K | Input— | Output— | Providers | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date2026-02-23 |
|
| DeepSeek V3.2DeepSeek | Context163.8K | Input$0.280/M | Output$0.420/M | Providers +170 | 39.5±6.9 | 51.6±8.6 | 51.4±8.7 | 40.6±16.0P | 49±16.1P | — | — | 46.2±16.0P | Throughput 46 t/s | Latency 5.36s | Release date2025-12-01 |
|
| DeepSeek V3.2 ExpDeepSeek | Context163.8K | Input$0.280/M | Output$0.420/M | Providers | — | 54±13.9P | 54.3±11.1E | — | 52.7±16.1P | — | — | — | Throughput — | Latency — | Release date2025-09-29 |
|
| DeepSeek V3.1 TerminusDeepSeek | Context131.1K | Input$0.270/M | Output$1.00/M | Providers +83 | — | 55.5±13.9P | 54.5±11.1E | — | 53±16.1P | — | — | — | Throughput 58 t/s | Latency 1.50s | Release date2025-09-22 |
|
| DeepSeek Chat V3.1DeepSeek | Context163.8K | Input— | Output— | Providers +6 | — | — | — | — | — | — | — | — | Throughput 23 t/s | Latency 2.02s | Release date2025-08-21 |
|
| DeepSeek R1DeepSeek | Context163.8K | Input$1.35/M | Output$4.20/M | Providers +143 | 41.4±16.0P | 50.3±13.9P | 51±8.7 | 45.7±16.0P | 57.1±11.8E | — | — | 43.3±16.0P | Throughput 52 t/s | Latency 10.36s | Release date2025-05-28 |
|
| Sonar Reasoning ProPerplexity | Context128K | Input— | Output— | Providers +3 | — | — | — | — | 57.4±11.8E | — | — | — | Throughput — | Latency — | Release date2025-03-07 |
|
| DeepSeek R1 Distill QwenDeepSeek | Context128K | Input— | Output— | Providers +50 | — | 40.9±13.9P | 39.6±8.9 | 55.6±16.0P | 55.7±11.8E | — | — | 37.3±16.0P | Throughput 50 t/s | Latency 19.88s | Release date2025-01-29 |
|
| R1 Distill Llama 70BDeepSeek | Context128K | Input$0.700/M | Output$1.05/M | Providers | — | 43.3±13.9P | 46±11.1E | — | 55±11.8E | — | — | — | Throughput — | Latency — | Release date2025-01-23 |
|
| DeepSeek V4DeepSeek | Context— | Input— | Output— | Providers +9 | — | — | — | — | — | — | — | — | Throughput 47 t/s | Latency 2.49s | Release date— |
|
| DeepSeek R1 Distill Qwen 1.5BDeepSeek | Context— | Input— | Output— | Providers +21 | — | 26.2±13.9P | 29.9±11.1E | — | 45.4±11.8E | — | — | — | Throughput 126 t/s | Latency 4.28s | Release date— |
|
| DeepSeek V2.5DeepSeek | Context— | Input— | Output— | Providers +3 | — | — | — | — | 45.4±16.0P | — | — | — | Throughput 16 t/s | Latency 0.96s | Release date— |
|
| DeepSeek V3.1DeepSeek | Context— | Input$0.560/M | Output$1.68/M | Providers +114 | 41.1±16.0P | 54.6±13.9P | 50±8.7 | 42.1±16.0P | 53±16.1P | — | — | 42.7±16.0P | Throughput 103 t/s | Latency 1.78s | Release date— |
|
| DeepSeek V3 TurboDeepSeek | Context— | Input— | Output— | Providers | — | — | — | — | — | — | — | — | Throughput 28 t/s | Latency 1.31s | Release date— |
|
| DeepSeek R1 TurboDeepSeek | Context— | Input— | Output— | Providers +1 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek Vision ReasonerDeepSeek | Context— | Input— | Output— | Providers | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek V3.2 ReasonerDeepSeek | Context— | Input— | Output— | Providers +1 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek Coder v2 InstructDeepSeek | Context— | Input— | Output— | Providers +2 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek Prover v2DeepSeek | Context— | Input— | Output— | Providers +6 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek Coder InstructDeepSeek | Context— | Input— | Output— | Providers +4 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek R1 MetaSearchDeepSeek | Context— | Input— | Output— | Providers | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek OCRDeepSeek | Context— | Input— | Output— | Providers +36 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| DeepSeek ReasonerDeepSeek | Context— | Input— | Output— | Providers +83 | — | — | — | — | — | — | — | — | Throughput 28 t/s | Latency 6.40s | Release date— |
|
| DeepSeek VL2 | Context— | Input— | Output— | Providers +8 | — | — | — | — | — | — | — | — | Throughput 126 t/s | Latency 0.75s | Release date— |
|
| DeepSeek Coder | Context— | Input— | Output— | Providers +8 | — | — | — | — | — | — | — | — | Throughput — | Latency — | Release date— |
|
| 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— |
|
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.
