DeepSeek V3.2 Speciale
Context 128K tokens
Top 1 cheapest providers (Adjusted $/task)
Within the active global provider, residency and confidentiality filters.
| # | Provider | Price channel | Raw input $/1M | Raw output $/1M | Adjusted $/task |
|---|---|---|---|---|---|
| 1 | Azure AI Foundrydirect — Direct price: Azure AI Foundry's own price list (https://prices.azure.com/api/retail/prices), collected 2026-11-03. | direct — Direct price: Azure AI Foundry's own price list (https://prices.azure.com/api/retail/prices), collected 2026-11-03. | $0.580 | $1.68 |
Composite
1 of 7 inputs6 radar axes: DesignArena's two boards share one17.3
Thin data · 1/7Based on only 1 of 7 Composite inputs — treat this rank as uncertaindominance-adjusted from 19.6: a better-measured model that is at least as good on each of these inputs ranks above it
AA Coding v4.x —Coding Agent v1.4 —AA Intelligence (est.) 6.2AA Agentic —Epoch ECI —Software ECI —DesignArena —/—
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
18 of 319 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
Agentic
Terminal-Bench Hard (AA)8034.8%
Terminal-Bench Hard (AA) pinned revision 74221fb → · Published board — Tests a pinned 44-task hard subset of terminal-based work using Terminus 2.
Instruction-following
IFBench (AA single-turn)7563.9%
IFBench (AA single-turn) published 2026-09-10 → · Published board — Tests precise single-turn instructions with deterministic rule checks.
Knowledge
Humanity's Last Exam (AA text-only)7928.7%
Humanity's Last Exam (AA text-only) published 2026-09-10 → · Published board — Tests expert-level knowledge on AA's text-only Humanity's Last Exam subset.
AA-Omniscience Index63-17.7
AA-Omniscience Index published 2026-09-10 → · Published board — Tests factual reliability while rewarding correct answers and penalizing hallucinations.
UGI NatInt8153.0
UGI NatInt published 2026-09-10 → · Published board — A private-question evaluation of general knowledge and reasoning across textbook, popular-culture and world-model topics.
Long-context
AA-LCR v1.16570.0%
AA-LCR v1.1 v1.1 → · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.
Math
AIME 2025 (AA)9996.7%
AIME 2025 (AA) v2025 → · Published board — Advanced mathematical problem solving on AIME I and II 2025.
HMMT November 2025 (MathArena) 2025-118993.3%
HMMT November 2025 (MathArena) 2025-11 → · Published board — The 30 final-answer problems of the November 2025 Harvard-MIT Mathematics Tournament, one of the largest high-school math competitions in the United States, run by MathArena on the contest problems.
Apex 2025 (MathArena)479.4%
Apex 2025 (MathArena) v2025 → · Published board — Twelve final-answer problems from 2025 competitions that MathArena selected because strong models of the time failed them; the selection is biased against the four models used to choose it (Grok 4, GPT-5, Gemini 2.5 Pro, GLM 4.5).
Apex Shortlist 2025 (MathArena)4269.7%
Apex Shortlist 2025 (MathArena) v2025 → · Published board — 47 final-answer problems from 2025 competitions on which at least one of two reference models (Grok 4 Fast, GPT-5 mini) got at least one of four attempts wrong, run by MathArena.
Science
CritPt (AA)807.4%
CritPt (AA) published 2026-09-10 → · Published board — Tests research-level physics reasoning with Python, symbolic and numerical answers.
GPQA Diamond (AA)8387.1%
GPQA Diamond (AA) published 2026-09-10 → · Published board — Tests graduate-level biology, physics and chemistry knowledge on the Diamond subset.
Tool-use
τ²-Bench Telecom (AA)30.0%
τ²-Bench Telecom (AA) published 2026-09-10 → · Published board — Tests dual-control telecom agents that coordinate tool use with a simulated user.
Uncensored
UGI Willingness764.80
UGI Willingness published 2026-09-10 → · Published board — A private-question evaluation of how far a model follows challenging instructions before refusing or deviating.
UGI10065.4
UGI published 2026-09-10 → · Published board — A private-question suite combining sensitive-topic knowledge with willingness to follow controversial instructions.
Writing
EQ-Bench Creative Writing v3511,277 Elo
EQ-Bench Creative Writing v3 v3 → · Published board — A LLM-judged creative writing benchmark with vocabulary and GPT-slop controls.
EQ-Bench Longform Creative Writing v1.115348.8
EQ-Bench Longform Creative Writing v1.11 → · Published board — LLM-judged benchmark in which models plan and write a short story/novella over 8x 1000-word turns, scored 0-100 across 14 rubric dimensions.
UGI Writing8354.0
UGI Writing published 2026-09-10 → · Published board — A maintainer-owned writing evaluation considering intelligence, style, repetition and length adherence, informed by human preferences.
Unusual results
2 threshold-crossing signals flagged · 15 eligible benchmark families
How flags are calculated
Heuristic screen, not statistical significance: benchmark families are correlated and source uncertainty is unknown. Peer evidence needs ≥ 20 independently measured matched configurations from ≥ 10 distinct model families. A flag needs a directed population z-score of magnitude ≥ 1.5 and a gap of ≥ 1.5 from the leave-one-benchmark-family-out mean z in the same direction, over ≥ 5 other benchmark families.
- unusually weak τ²-Bench Telecom (AA) published 2026-09-10
Why
Observed: 0 fractionPeer mean 0.51002 · peer sd 0.31633 · n 440 · families 323Directed z -1.612 · baseline z 0.837 · gap -2.449 · profile n 140 fractionmeasuredobserved 2026-09-10artificialanalysis.ai ↗Evidence
Axis: τ²-Bench Telecom (AA) · published 2026-09-10 · Published boardExact value:0fractionObserved: 2026-09-10T21:47:16.627Z · publication date: not recordedObservation id:aa:527e943a-adc6-4e69-93af-d1608e1b5fed:tau2 - unusually strong UGI published 2026-09-10
Why
Observed: 65.37 scorePeer mean 35.1234 · peer sd 14.30074 · n 100 · families 84Directed z 2.115 · baseline z 0.571 · gap 1.544 · profile n 1465.37 scoremeasuredobserved 2026-09-10huggingface.co ↗Evidence
Axis: UGI · published 2026-09-10 · Published boardExact value:65.37scoreObserved: 2026-09-10T22:04:13Z · publication date: not recordedObservation id:public:bace046d11dcb7a0955612f9
Missing coverage · 301 benchmark versions
No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.
Token offers by platform · 1 offer (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
1 offer within the active global filters; “price n/a” means the provider serves the model but publishes no direct token price we could collect — no other channel's price is shown in its place.
Azure AI Foundry (1)
Direct prices from Azure AI Foundry's own price list. Source: prices.azure.com/api/retail/prices, collected 2026-11-03.
| Azure AI Foundry | global | $0.580 raw in $/1M | $1.68 raw out $/1M |