Sapiens AIAgnes 2.5 Pro Theta
2 cheaper ranked models score higher: GPT-5 nano (55.9, $0.138), Ling 3.0 Flash (51.8, $0.115).
As of 9 Nov 2026, Agnes 2.5 Pro Theta by Sapiens AI scores 50.8 on the BenchLeader Index (599 ranked configurations), with its best results in long context (67) and weakest in knowledge (39). API pricing: $0.450 per 1M input tokens and $0.900 per 1M output tokens ($0.563 blended at 3:1), list price on 9 Nov 2026. It streams about 210 tokens/s (Artificial Analysis). Released 7 Nov 2026.
Key facts (as of 9 Nov 2026)
- Agnes 2.5 Pro Theta API pricing: $0.450 per 1M input tokens and $0.900 per 1M output tokens ($0.563 blended at 3:1) (list price, 9 Nov 2026).
- Agnes 2.5 Pro Theta streams about 210 tokens/s (Artificial Analysis, 9 Nov 2026).
- Agnes 2.5 Pro Theta was released on 7 Nov 2026 (Sapiens AI).
In detail. Agnes 2.5 Pro Theta is a Sapiens AI proprietary reasoning model, released 7 Nov 2026. Its best configuration ranks #327 of 599 on the BenchLeader Index at 50.8 ±8.2, in the lower half. The ± is the point: 378 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in long context (67) and lowest in knowledge (39). At $0.150 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 212 tokens per second puts it in the fastest quarter, with a first answer in 20.9 s. Last measured 7 Nov 2026.
- Blended price
- $0.150/M
- $0.100 in · $0.300 out
- via Artificial Analysis
- Output speed
- 212 tok/s
- measured by Artificial Analysis
- First answer
- 21 s
- Context
- 1M
- via Artificial Analysis
- Full answer
- 30 s
- median, reasoning included
- Cost per task
- $0.575
- 22 min a task, AA Index
- Tokens per task
- 278k
- 83% of it reasoning
- Released
- 7 Nov 2026
- Overall index51
- Reasoning44
- Coding46
- Knowledge39
- Multimodal61
- Long context67
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source |
|---|---|---|
| Humanity's Last Exam (AA)not in index | 37.2%#118 | Artificial Analysis |
| HLE-Diamond (AA) | 16.8%#86 | Artificial Analysis |
| CritPtnot in index | 15.1%#102 | Artificial Analysis |
| CritPt v2 hard subset (AA) | 26.0%#82 | Artificial Analysis |
Coding
| Benchmark | not stated | Source |
|---|---|---|
| SciCode (AA)not in index | 56.0%#49 | Artificial Analysis |
| Terminal-Bench 4.0 (AA)not in index | 10.1%#112 | Artificial Analysis |
| FrontierCode 2.0 (AA) | 16.6%#74 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source |
|---|---|---|
| GDPval-AA v2.1not in index | 1507#41 | Artificial Analysis |
| AutomationBenchnot in index | 48.8%#104 | Artificial Analysis |
| GDP.pdf (AA)not in index | 14.2%#123 | Artificial Analysis |
| AA-Briefcase v1.1not in index | 1496#44 | Artificial Analysis |
| Terminal-Bench-Science 0.1 (AA)not in index | 3.8%#84 | Artificial Analysis |
Knowledge
| Benchmark | not stated | Source |
|---|---|---|
| AA-Omniscience v1.1 | -11.2#101 | Artificial Analysis |
| AA-Omniscience v1.1: accuracynot in index | 21.3%#109 | Artificial Analysis |
| AA-Omniscience v1.1: non-hallucinationnot in index | 58.6%#56 | Artificial Analysis |
Multimodal
| Benchmark | not stated | Source |
|---|---|---|
| MMMU-Pro | 78.4%#84 | Artificial Analysis |
Long context
| Benchmark | not stated | Source |
|---|---|---|
| AA-LCR v1.1 | 84.7%#6 | Artificial Analysis |
Composite
| Benchmark | not stated | Source |
|---|---|---|
| AA Intelligence Index v5.0not in index | 24.0#77 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | – | 22.3 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0014 | – | 23.7 s |
| Code edit | 6,000 / 1,500 | $0.0011 | – | 28.0 s |
| Agentic coding session | 60,000 / 4,000 | $0.0072 | – | 39.7 s |
| Structured extraction | 2,000 / 200 | $0.0003 | – | 21.8 s |
More from Sapiens AI
Compare Agnes 2.5 Pro Theta with…
See also
Embed or cite
Show where Agnes 2.5 Pro Theta ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.
[](https://www.benchleader.com/models/agnes-2-5-pro-theta)Cite this
The ranking changes daily, so a reference carries the data date. APA style:
BenchLeader. (2026, November 9). Agnes 2.5 Pro Theta on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 9, 2026, from https://www.benchleader.com/models/agnes-2-5-pro-theta
BibTeX:
@misc{benchleader_agnes_2_5_pro_theta_20261109,
author = {{BenchLeader}},
title = {{Agnes 2.5 Pro Theta on the BenchLeader Index (method v1.4)}},
year = {2026},
month = nov,
howpublished = {\url{https://www.benchleader.com/models/agnes-2-5-pro-theta}},
note = {Data as of 2026-11-09}
}Every badge style, the embeddable top-10 widget and how to cite the Index
Questions people ask
- What are Agnes 2.5 Pro Theta's benchmark scores?
- As of 9 Nov 2026, Agnes 2.5 Pro Theta scores 50.8 ±8.2 on the BenchLeader Index, strongest in long context (67) and weakest in knowledge (39). That rests on independent benchmark results only, and every score on this page links to its publisher.
- How does Agnes 2.5 Pro Theta rank?
- Agnes 2.5 Pro Theta is #327 of 599 ranked configurations on 9 Nov 2026. 378 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 69.4.
- How much does Agnes 2.5 Pro Theta cost per 1M tokens?
- As of 9 Nov 2026, Agnes 2.5 Pro Theta costs $0.100 per 1M input tokens and $0.300 per 1M output tokens, $0.150 blended (three input tokens per output token).
- Is Agnes 2.5 Pro Theta better than Ling 3.1 Flash?
- As of 9 Nov 2026, Agnes 2.5 Pro Theta and Ling 3.1 Flash are neck and neck on quality (50.8 vs 50.8 on the BenchLeader Index, inside the ±10.9 margin of error), and Agnes 2.5 Pro Theta costs 3.0× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.
- When was Agnes 2.5 Pro Theta released?
- Sapiens AI released Agnes 2.5 Pro Theta on 7 Nov 2026.
Data as of 9 Nov 2026. Compare with another model.
Cite as: BenchLeader, “Agnes 2.5 Pro Theta: benchmarks, pricing, speed and rank”, data as of 9 Nov 2026, https://www.benchleader.com/models/agnes-2-5-pro-theta.