stealthmodels.

MODEL BENCHMARKS

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

Pareto 26.9, first seen as Union Alpha, combines multiple models in one hosted system. This record retains that version's identity and benchmark history.

BENCHMARK BY BENCHMARK

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) benchmark scores

StealthMark
62.6/100Overall rank 8/58
General Intelligence
58.7/100
Visual Intelligence
69.6/100

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)Other modelsMedian

Cost efficiency

How much Performance you get for your money.

Rank 30/52

DATASHEET

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) specs

A good fit forComparing the 26.9 system's cost and capability with individual models. Check the selected API version because the unversioned route has advanced.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)Other modelsMedian

Context windowTokens in one request
262,000Larger than 1 of 49 others
Output speedTokens per second, our measurement
26 tokens/sSlowest of 52 models
Developer
Circuit & Chisel / Unbiased
Released
20 Sep 2026
Family
Pareto
Open model
No
Local model
No
Origin
United States
Parameters
Not disclosed
Max output
131,000 tokens
Input → output
Text, image → text
All prices and providers

Pareto 26.9 historical rates: USD $2.50 input, $0.25 cached input and $7.50 output per 1M tokens. The unversioned route now serves 26.10 Preview.

  • UnbiasedObserved 2026-10-06
    Input $2.5Output $7.5Cached read $0.25

    per 1M tokens

    Historical rates

    unbiased/pareto
Measurements and sources6

Our speed measurement: 25.74 tokens/s on average, from 21.62 to 29.47 tokens/s.

GAME BENCHMARKS

Watch Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) play

Foresight

Anticipating threats and planning ahead

78.7/100Rank 1/54
Survived 2 of 3
Foresight: complete playing field
Level 319 ptsTurn 20 / 20

Points per level

  1. Level 1Second Thoughts27 ptsBest 27Survived
  2. Level 2Pressure Pot12 ptsBest 13Out of lives
  3. Level 3Hide or Flight19 ptsBest 19Survived

Collect coins while avoiding ghosts. Ghosts move before the player each turn, and collisions cost a life. 3 levels in sequence, up to 66 turns in total. Earlier moves and outcomes carry forward, testing learning across levels.

Latch

Cooperative planning and learning from experience

36.5/100Rank 37/55
Cleared 4 of 5
Latch: complete playing field
Level 519.4 ptsTurn 33 / 40

Points per level

  1. Level 1Hold the door5.2 ptsBest 5.2Cleared
  2. Level 2Crossfeed9.8 ptsBest 10.2Cleared
  3. Level 3Relay vault14.2 ptsBest 19.2ClearedFinished early · 4 beacons left
  4. Level 4Passing bay9.2 ptsBest 18.2Finished early
  5. Level 5Second thoughts19.4 ptsBest 20.8Cleared

Coordinate two robots to push numbered cores onto matching sockets. Pressure plates hold doors open; moving either robot uses one turn. 5 levels in sequence, up to 152 turns in total. Earlier moves and outcomes carry forward, testing learning across levels.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) benchmark results ↗

StealthMark score

StealthMark measures how well a model’s abilities carry over to original, demanding tasks it could not have trained for: physical and spatial reasoning, planning across many steps, and precise visual work. How StealthMark works explains every benchmark, what it tests and why it is hard.

General Intelligence

The ability to reason, plan, solve problems and understand how things relate and behave. A high score means it thinks hard problems through, from careful analysis to plans that span many steps.

Visual Intelligence

The ability to understand visual intent and turn ideas, descriptions and plans into coherent, effective visuals. A high score means design ideas come out the way you meant them, in illustrations, icons, diagrams or interface visuals.

Overall score

Overall shows how well a model’s abilities carry over to original, demanding tasks, weighting every benchmark into one score. Estimated means the total includes an estimate for one missing scene.

Judging

Scores reflect how well each result meets the benchmark requirements. Game scores come from the recorded play: the points earned on each level and how each level ended. How StealthMark keeps scores honest.

Games

Each game is a sequence of levels, and earlier moves carry into later levels. A level that was not played is shown as not played and never counts as zero. Play every level of Latch or Foresight yourself to get a score calculated like the models’.

Reasoning efficiency

Good results with little thinking score high. On each task, a model’s result and reasoning tokens are compared with every other model and combined with the Balanced Integration Score (Liesefeld & Janczyk, 2019); 50 is average. Single step shows how much reasoning a model needs to understand and solve one problem, multi-step how efficiently it keeps working on a task by building on its earlier reasoning, as in agentic workflows. Not part of StealthMark.

Cost efficiency

How much StealthMark a model delivers for its money, in StealthMark points per dollar: its overall score divided by what one benchmark task costs it on average. The cost is what our runs were charged, or their tokens at the model’s list price; benchmarks a model has not run yet are estimated from its other costs. Like StealthMark, it is absolute and compares with no other model. Not part of StealthMark.

BENCHMARK RESULTS

Compare results

This link opens the same results in the same order.

StealthMark benchmark results

Reset benchmark preferences?

Return to the default view, clear comparison selections and restore the advertisement.