ExamenosParticle-1
Explore Particle-1 from Particle AI: published specifications, source-linked vendor benchmarks, independent evaluator coverage and recorded pricing when available.
Compare Particle-1 with other models →Compare subscription plansExplore data coverage
Pricing · OpenRouter
Dated cached OpenRouter rates in USD per 1M tokens. Open the dashboard for live enhancements. Per-metric endpoint minima can refer to different providers; they are not a guaranteed combined rate from one endpoint.
| Tier | Input / 1M | Output / 1M | Cached input / 1M | Cache write / 1M | Date & source |
|---|---|---|---|---|---|
| Default | $0.05 | $0.5 | $0.003 | — | 2026-11-01 · OpenRouter source |
Recorded pricing notes
OpenRouter list rate captured 1 November 2026, pending the next min-healthy refresh. No contributor sibling. OpenRouter canonical particle-ai/particle-1-20261027. Particle docs state $0.25 input, $0.03 cached and $1.00 output; its launch footnote uses $0.30/$0.006/$1.20 as evaluation prices.
Official / vendor benchmarks
Default headline records. Own-vendor, peer-vendor and third-party provenance remain visible in evidence; configurations may differ.
| Benchmark / evaluator | Headline score | Evidence |
|---|---|---|
| DeepSWE 1.1 pass rate (Particle launch) | 72.6%max effort | All 1 recorded result & sources72.6% · raw 72.57 % Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. 1304 of 1797 passed. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| DeepSWE 1.1 cost per task (Particle launch) | $0.221/taskmax effort | All 1 recorded result & sources$0.221/task · raw 0.2211 USD/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Evaluation at the $0.30 input, $0.006 cached and $1.20 output footnote prices, fixed 95% cache hit, all attempts; not current list prices. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| DeepSWE 1.1 mean output tokens (Particle launch) | 59580 tokens/taskmax effort | All 1 recorded result & sources59580 tokens/task · raw 59580 tokens/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| Terminal-Bench 2.1, excluding fastText pass rate (Particle launch) | 79.9%max effort | All 1 recorded result & sources79.9% · raw 79.93 % Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. 1099 of 1375 passed. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| Terminal-Bench 2.1, excluding fastText cost per task (Particle launch) | $0.049/taskmax effort | All 1 recorded result & sources$0.049/task · raw 0.0489 USD/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Evaluation at the $0.30 input, $0.006 cached and $1.20 output footnote prices, fixed 95% cache hit, all attempts; not current list prices. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| Terminal-Bench 2.1, excluding fastText mean output tokens (Particle launch) | 24594 tokens/taskmax effort | All 1 recorded result & sources24594 tokens/task · raw 24594 tokens/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ReactBench pass rate (Particle launch) | 31.5%max effort | All 1 recorded result & sources31.5% · raw 31.46 % Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. 196 of 623 passed. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ReactBench cost per task (Particle launch) | $0.095/taskmax effort | All 1 recorded result & sources$0.095/task · raw 0.0947 USD/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Evaluation at the $0.30 input, $0.006 cached and $1.20 output footnote prices, fixed 95% cache hit, all attempts; not current list prices. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ReactBench mean output tokens (Particle launch) | 41314 tokens/taskmax effort | All 1 recorded result & sources41314 tokens/task · raw 41314 tokens/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ADE Bench pass rate (Particle launch) | 64.5%max effort | All 1 recorded result & sources64.5% · raw 64.45 % Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. 495 of 768 passed. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ADE Bench cost per task (Particle launch) | $0.016/taskmax effort | All 1 recorded result & sources$0.016/task · raw 0.0162 USD/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Evaluation at the $0.30 input, $0.006 cached and $1.20 output footnote prices, fixed 95% cache hit, all attempts; not current list prices. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| ADE Bench mean output tokens (Particle launch) | 9645 tokens/taskmax effort | All 1 recorded result & sources9645 tokens/task · raw 9645 tokens/task Headline · max effort · Own vendor Source/record date: 2026-10-28 Harness: Terminus 2. Particle launch post value. The embedded data object behind this chart is labelled draft_with_mock_values, with simulated accuracy and token sources, although the post presents it as a result, so treat it as an unverified vendor statement. Setup: Terminus 2 agent, 512k-token limit, temperature 1, top_p 0.95, max reasoning, 16 rollouts per task, infrastructure-interrupted rollouts excluded. https://particle.ai/blog/introducing-particle-1 |
| GPQA Diamond | 90.4%unknown effort | All 1 recorded result & sources90.4% · raw 90.4 % Headline · unknown effort · Own vendor Source/record date: 2026-10-28 Particle launch post prose; general benchmark outside its training domains. Harness and evaluation date unpublished. The post reports base DeepSeek V4.1 Flash at 90.91. https://particle.ai/blog/introducing-particle-1 |
| BabyVision (with tools) | 89.2%unknown effort | All 1 recorded result & sources89.2% · raw 89.18 % Headline · unknown effort · Own vendor Source/record date: 2026-10-28 Particle launch post prose; general benchmark outside its training domains. Harness and evaluation date unpublished. The post reports base DeepSeek V4.1 Flash at 89.60. https://particle.ai/blog/introducing-particle-1 |
| Tau Airlines (Particle launch; version unstated) | 86%unknown effort | All 1 recorded result & sources86% · raw 86 % Headline · unknown effort · Own vendor Source/record date: 2026-10-28 Particle launch post prose; general benchmark outside its training domains. Harness and evaluation date unpublished. The post reports base DeepSeek V4.1 Flash at 88.00. https://particle.ai/blog/introducing-particle-1 |
| Standard MMLU (Particle launch) | 86.1%unknown effort | All 1 recorded result & sources86.1% · raw 86.08 % Headline · unknown effort · Own vendor Source/record date: 2026-10-28 Particle launch post prose; general benchmark outside its training domains. Harness and evaluation date unpublished. The post reports base DeepSeek V4.1 Flash at 85.98. https://particle.ai/blog/introducing-particle-1 |
Independent evaluators
Evaluator harnesses are distinct from vendor measurements. Missing coverage is not a failed test.
| Benchmark / evaluator | Headline score | Evidence |
|---|---|---|
| DataAnalyticsBench · questions correct /43 · Plotly DataAnalyticsBench | 40 correctunknown effort | All 1 recorded result & sources40 correct · raw 40 correct Headline · unknown effort · Independent evaluator Source/record date: 2026-11-01 Harness: Plotly Studio Embedded, one run on 43 private English data-analytics questions in a synthetic wind-energy warehouse. Claude Sonnet 5 judges against Plotly answer key. One or two questions can be noise; no error bars. Exact checkpoint hash, run date and effort unpublished. Company-operated analytics benchmark, distinct from provider results. Plotly reports $0.35 a run and calls it a fine-tune of DeepSeek V4.1 Flash; deep dive https://plotly.com/blog/particle-1-plotly-data-analytics-bench/ on 31 October. https://plotly.com/blog/introducing-dataanalyticsbench/ |
Read how we select and source scores or the comparison guide.