Capability
Unranked
field median 56.3
Not eligible for a public rank
Pareto 26.9
Decision reading
Pareto 26.9 is tracked, but not publicly ranked yet. The profile exposes 4 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Sep 17, 2026 — see all recent releases
Data as of September 18, 2026 · How the score is built
Pareto 26.9 will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Share or export
Strongest published evidence
Published rows are visible, but no category has enough eligible evidence for a comparative rank.
Validate before choosing
4 published rows leave some tracked benchmark slots empty. Independent runtime speed has not been measured.
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Price
$2.50input / $7.50 output
input median $0.95
cached $0.25 · blended $5
Speed
Not measured
field median 90 tok/s
Time to first token not measured
Context
N/A
field median 256,000
Reported for this model; direct source link not stored
How much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic1/1 verified
- Coding1/1 verified
- ReasoningNot measured
- Knowledge1/1 verified
- MathNot measured
- MultilingualNot measured
- Multimodal1/1 verified
- Inst. FollowingNot measured
Verified sourceProvisionalNot measured
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%1 benchmarkVerified | Score pending | Not ranked | Not available | 22% | 1 benchmark | Verified |
| CodingWeight 20%1 benchmarkVerified | Score pending | Not ranked | Not available | 20% | 1 benchmark | Verified |
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | Not ranked | Not available | 17% | 0 benchmarks | Not measured |
| KnowledgeRank Not rankedWeight 12%1 benchmarkVerified | 77.4 | Not ranked | Not available | 12% | 1 benchmark | Verified |
| MathWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalRank Not rankedWeight 12%1 benchmarkVerified | 60.9 | Not ranked | Not available | 12% | 1 benchmark | Verified |
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | Not ranked | Not available | 5% | 0 benchmarks | Not measured |
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| DeepSWE | Score74.0% | Versus best verified row Best verified: Muse Spark 1.3 · 75.4% | Gap1.4 behind | WeightDisplay only | Provider exact |
Agentic1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 4.0 | Score51.00% | Versus best verified row Best verified: Claude Mythos 5.1 · 60.90% | Gap9.9 behind | WeightDisplay only | Provider exact |
Knowledge1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HLE w/o toolsHumanity's Last Exam without tools | Score49% | Versus best verified row Best verified: Claude Fable 5.1 · 60.9% | Gap11.9 behind | WeightDisplay only | Provider exact |
Multimodal1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMMU-ProMassive Multi-discipline Multimodal Understanding Pro | Score78% | Versus best verified row Best verified: Gemini 3.5 Flash · 83.6% | Gap5.6 behind | WeightDisplay only | Provider exact |
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- paretoUnbiased pricing
- Context window
- N/A
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not disclosed by the provider
- Availability
- Sold only as the `pareto` model string on Unbiased's OpenAI-compatible API at https://api.unbiased.ai/v1, which also accepts Anthropic /v1/messages traffic. Access is prepaid pay-as-you-go credits and the service is in beta with every signup reviewed by hand; GET /v1/models is not implemented. Text and image input, text output. Unbiased publishes no context-window figure and no weights.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.25 per million cached input tokensUnbiased pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
Lineage
The sequence follows explicit supersedes links. A successor's displayed score stays at least 0.1 points above its predecessor; raw benchmark rows do not move. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Sep 17, 2026 · you are here
Pareto 26.9Not publicly ranked · $2.5 / $7.5
Base entry
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Pareto 26.9, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
Pareto 26.9 is a proprietary model with a N/A context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Sold only as the `pareto` model string on Unbiased's OpenAI-compatible API at https://api.unbiased.ai/v1, which also accepts Anthropic /v1/messages traffic. Access is prepaid pay-as-you-go credits and the service is in beta with every signup reviewed by hand; GET /v1/models is not implemented. Text and image input, text output. Unbiased publishes no context-window figure and no weights.
Pareto is not a single model. Unbiased describes it as a blend that engages multiple LLMs in parallel on every request and synthesizes one answer, drawing on both frontier and open models, and states that the composition can change between releases. Scores therefore cannot be attributed to any one model, and every row here is blocklisted from weighted ranking. The four stored values are the Pareto column of Unbiased's own September 2026 model-card table (DeepSWE 74, Terminal-Bench 4.0 51, MMMU-Pro 78, HLE no-tools 49), produced on the public MIT-licensed pareto-evals harness. Unbiased also publishes ArXivMath 88, which BenchLM does not store: that harness reads the May 2026 MathArena ArXiv set, a different release from the June 2026 ArXivMath lane the catalog tracks. Unbiased states that measured task costs and a composite score are not published for this release. BenchLM does not republish the competitor column of that table.
4 of 446 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Radar
Full release historyPareto 26.9 release history
Pareto 26.9 released
Unbiased · Model release
Source confirmed
Radar confirmed these at the source. Use Pareto 26.9 in your work? Explore Radar to follow supported changes and choose your alerts.
Questions
How does Pareto 26.9 perform overall in AI benchmarks?
Pareto 26.9 has 4 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Is Pareto 26.9 good for knowledge and understanding?
Pareto 26.9 has source-displayable benchmark coverage for knowledge and understanding, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Pareto 26.9 good for coding and programming?
Pareto 26.9 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Pareto 26.9 good for agentic tool use and computer tasks?
Pareto 26.9 has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Pareto 26.9 good for multimodal and grounded tasks?
Pareto 26.9 has source-displayable benchmark coverage for multimodal and grounded tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Does Pareto 26.9 have full benchmark coverage on BenchLM?
No. Pareto 26.9 currently has 4 source-displayable rows across 446 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Pareto 26.9?
Pareto 26.9 has a reported context window of N/A in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.
Compare Pareto 26.9 with every tracked model489 comparisons
Last updated September 18, 2026. Runtime fields remain blank until a sourced snapshot exists.
Watch Pareto 26.9 in the weekly brief
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.