Models · OpenAI · Out sinceReleased 29 Sep 2026

GPT-6.1 Sol#4 for coding.#4 for coding, best at max effort.

Available on OpenRouterOn OpenRouterAvailable nowGenerally availableClosed (can't be downloaded)Proprietary

Where you can use it:In apps:ChatGPT (Business plan) · GPT-6.1 Sol (Work/Codex only)

GPT-6.1 Sol is made by OpenAI. Among the models we track it ranks #4 for coding, #11 for research and analysis. It's mid-priced to use.

Default reasoning.effort medium; none/minimal not supported. Cached input $0.10/M (95% off). Max input 922K. Prompts >272K input billed 2x input / 1.5x output. reasoning.mode "pro" available (see -pro variant). Knowledge cutoff Apr 30, 2026. OpenAI Codex docs recommend it for complex coding and agentic workflows.

Writing & creativity: —
Research & analysis: 73.8 / 100 · #11
Coding: 83.7 / 100 · #4
Price: Mid-priced$2 / $10
Price per 1M (blended)Blended / 1M: $4
MemoryContext: 1.1M
Longest answerMax output: 128K
Test resultsResults: 120 (95 independent95 indep.)
Out sinceReleased: 29 Sep 2026
Made byVendor: OpenAI
UnderstandsInputs: text, image

Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.

Thinking levels

Reasoning settings

LowMediumHighExtra highMax

Highlighted: where it did best for coding (Maximum thinking).

Highlighted: dominant setting in its coding composite.

Links

Using it through OpenRouter

On OpenRouter

OpenRouter is a service that gives access to many AI models in one place. This model is listed there as openai/gpt-6.1-sol.

Route it as openai/gpt-6.1-sol at $2 in / $10 out per 1M tokens, 1.1M context. Listed since 29 Sep 2026.

include_reasoningmax_completion_tokensmax_tokensreasoningreasoning_effortresponse_formatseedstructured_outputstool_choicetools

Thinking level

Reasoning effort

How long should GPT-6.1 SolWhere GPT-6.1 Solthink?peaks.

Each chart is one test. Left to right, the model thinks longer; higher is a better result. The yellow ring marks where it did best. Thinking level is how long the AI thinks before it answers. Higher is slower and costs more, and it isn't always better.

Score per reasoning setting on each benchmark with ≥2 settings; ring = peak. Canonical results (independent over vendor).

Artificial Analysis Coding Agent Index

Best at Extra high, worse abovePeaks at Extra high

Terminal-Bench 4.0

Best at MaxBest at Max

DeepSWE

Best at Extra high, worse abovePeaks at Extra high

SWE Atlas - Codebase QnA

Best at Extra high, worse abovePeaks at Extra high

SciCode

Best at High, worse abovePeaks at High

Artificial Analysis Intelligence Index

Best at MaxBest at Max

Terminal-Bench-Science 0.1

Best at MaxBest at Max

AutomationBench

Best at MaxBest at Max

OSWorld 2.0 offline set

Best at MaxBest at Max

AA-LCR

Best at Low, worse abovePeaks at Low

AA-Omniscience Accuracy

Best at MaxBest at Max

AA-Omniscience Hallucination Rate

Best at High, worse abovePeaks at High

Lower is better on this test

AA-Omniscience Index

Best at MaxBest at Max

ARC-AGI-2

Best at MaxBest at Max

ARC-AGI-3

Best at Extra high, worse abovePeaks at Extra high

GDP.pdf

Best at High, worse abovePeaks at High

Humanity's Last Exam

Best at MaxBest at Max

Every test result

Every result

All the numbers,with where they came from.

Every result we've found for this model, with who measured it and a link to where we read it.

All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.

All results for this model
						NotesNotes
AA-Briefcase v1.1	1564	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-LCR	84.0%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-LCR accuracy, AA-run (long multi-document reasoning)
AA-LCR	83.3%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-LCR accuracy, AA-run (long multi-document reasoning)
AA-LCR	82.3%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-LCR accuracy, AA-run (long multi-document reasoning)
AA-LCR	79.7%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-LCR accuracy, AA-run (long multi-document reasoning)
AA-LCR	83.0%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-LCR accuracy, AA-run (long multi-document reasoning)
AA-Omniscience Accuracy	58.9%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-Omniscience Accuracy	60.4%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-Omniscience Accuracy	60.8%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-Omniscience Accuracy	60.8%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-Omniscience Accuracy	62.1%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—
AA-Omniscience Hallucination Rate	51.6%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page)
AA-Omniscience Hallucination Rate	51.6%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page)
AA-Omniscience Hallucination Rate	49.4%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page)
AA-Omniscience Hallucination Rate	50.9%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page)
AA-Omniscience Hallucination Rate	54.3%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	share of non-correct answers that were wrong instead of abstaining; = 1 - AA 'omniscienceNonHallucination' field (equals breakdown.hallucinationRate on the eval page)
AA-Omniscience Index	37.6	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised
AA-Omniscience Index	40	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised
AA-Omniscience Index	41.5	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised
AA-Omniscience Index	40.9	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised
AA-Omniscience Index	41.5	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-Omniscience Index (-100..100): correct minus incorrect, abstentions not penalised
ARC-AGI-2	86.7%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; $0.10/task; data https://arcprize.org/media/data/leaderboard/v2.json
ARC-AGI-2	91.7%	HighGPT-6.1 Sol (High)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; $0.13/task; data https://arcprize.org/media/data/leaderboard/v2.json
ARC-AGI-2	91.7%	Extra highGPT-6.1 Sol (XHigh)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; $0.18/task; data https://arcprize.org/media/data/leaderboard/v2.json
ARC-AGI-2	94.2%	MaxGPT-6.1 Sol (Max)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; $0.25/task; data https://arcprize.org/media/data/leaderboard/v2.json
ARC-AGI-3	82.8%	LowGPT-6.1 Sol - Provider Adapter (Low)	Independent testIndependentARC Prize ↗	30 Sep 2026	Provider Adapter (vendor-built agent harness)	semi-private set; total cost $6,622; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	3.9%	LowGPT-6.1 Sol (Low)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; total cost $7,111; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	91.0%	MediumGPT-6.1 Sol - Provider Adapter (Medium)	Independent testIndependentARC Prize ↗	30 Sep 2026	Provider Adapter (vendor-built agent harness)	semi-private set; total cost $5,835; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	10.6%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; total cost $7,370; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	95.0%	HighGPT-6.1 Sol - Provider Adapter (High)	Independent testIndependentARC Prize ↗	30 Sep 2026	Provider Adapter (vendor-built agent harness)	semi-private set; total cost $5,404; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	26.7%	HighGPT-6.1 Sol (High)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; total cost $9,758; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	96.4%	Extra highGPT-6.1 Sol - Provider Adapter (XHigh)	Independent testIndependentARC Prize ↗	30 Sep 2026	Provider Adapter (vendor-built agent harness)	semi-private set; total cost $4,360; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	39.9%	Extra highGPT-6.1 Sol (XHigh)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; total cost $9,104; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	96.2%	MaxGPT-6.1 Sol - Provider Adapter (Max)	Independent testIndependentARC Prize ↗	30 Sep 2026	Provider Adapter (vendor-built agent harness)	semi-private set; total cost $3,817; data https://arcprize.org/media/data/leaderboard/v3.json
ARC-AGI-3	52.7%	MaxGPT-6.1 Sol (Max)	Independent testIndependentARC Prize ↗	30 Sep 2026	—	semi-private set; total cost $7,632; data https://arcprize.org/media/data/leaderboard/v3.json
Artificial Analysis Coding Agent Index	57.2	LowGPT-6.1 Sol (low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	agent Codex; components: DeepSWE v1.1 67.6, SWE-Atlas-QnA 55.1, Terminal-Bench v4 49.0; avg cost $0.50/task; avg wall time 9 min/task
Artificial Analysis Coding Agent Index	61.4	MediumGPT-6.1 Sol (medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	agent Codex; components: DeepSWE v1.1 72.0, SWE-Atlas-QnA 60.8, Terminal-Bench v4 51.5; avg cost $0.70/task; avg wall time 11 min/task
Artificial Analysis Coding Agent Index	60.1	HighGPT-6.1 Sol (high)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	agent Codex; components: DeepSWE v1.1 70.5, SWE-Atlas-QnA 59.9, Terminal-Bench v4 50.0; avg cost $0.89/task; avg wall time 13 min/task
Artificial Analysis Coding Agent Index	62.9	Extra highGPT-6.1 Sol (xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	agent Codex; components: DeepSWE v1.1 73.2, SWE-Atlas-QnA 61.0, Terminal-Bench v4 54.5; avg cost $1.04/task; avg wall time 16 min/task
Artificial Analysis Coding Agent Index	60.1	MaxGPT-6.1 Sol (max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	agent Codex; components: DeepSWE v1.1 69.6, SWE-Atlas-QnA 57.8, Terminal-Bench v4 53.0; avg cost $1.55/task; avg wall time 24 min/task
Artificial Analysis Intelligence Index	42.1	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA slug gpt-6-1-sol-low; list price $2/10 per 1M in/out; cost to run AA Intelligence Index $0.13/task
Artificial Analysis Intelligence Index	47.8	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA slug gpt-6-1-sol-medium; list price $2/10 per 1M in/out; cost to run AA Intelligence Index $0.21/task
Artificial Analysis Intelligence Index	50.2	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA slug gpt-6-1-sol-high; list price $2/10 per 1M in/out; cost to run AA Intelligence Index $0.32/task
Artificial Analysis Intelligence Index	51	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA slug gpt-6-1-sol-xhigh; list price $2/10 per 1M in/out; cost to run AA Intelligence Index $0.39/task
Artificial Analysis Intelligence Index	51.8	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA slug gpt-6-1-sol; list price $2/10 per 1M in/out; cost to run AA Intelligence Index $0.72/task
Artificial Analysis output speed	55 tok/s	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	tokens/sec (median output speed, first-party API); TTFT 2.7s; list price $2/10 per 1M in/out
Artificial Analysis output speed	58 tok/s	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	tokens/sec (median output speed, first-party API); TTFT 5.7s; list price $2/10 per 1M in/out
Artificial Analysis output speed	60 tok/s	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	tokens/sec (median output speed, first-party API); TTFT 58.0s; list price $2/10 per 1M in/out
Artificial Analysis output speed	63 tok/s	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	tokens/sec (median output speed, first-party API); TTFT 108.9s; list price $2/10 per 1M in/out
Artificial Analysis output speed	64 tok/s	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	tokens/sec (median output speed, first-party API); TTFT 293.4s; list price $2/10 per 1M in/out
AutomationBench	24.7%	Lowreasoning effort=low	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	AutomationBench 1.0.6 (Zapier); vendor-estimated API cost/task $0.157
AutomationBench	31.7%	Mediumreasoning effort=medium	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	AutomationBench 1.0.6 (Zapier); vendor-estimated API cost/task $0.1917
AutomationBench	33.2%	Highreasoning effort=high	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	AutomationBench 1.0.6 (Zapier); vendor-estimated API cost/task $0.2255
AutomationBench	35.5%	Extra highreasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	AutomationBench 1.0.6 (Zapier); vendor-estimated API cost/task $0.2508
AutomationBench	36.1%	Maxreasoning effort=max	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	AutomationBench 1.0.6 (Zapier); vendor-estimated API cost/task $0.2989
DeepSWE	67.6%	LowGPT-6.1 Sol (low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
DeepSWE	64.4%	Lowreasoning effort=low	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	DeepSWE v1.1 (113 long-horizon SWE tasks); GPT models run in Codex; vendor-estimated API cost/task $0.1714
DeepSWE	72.0%	MediumGPT-6.1 Sol (medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
DeepSWE	73.0%	Mediumreasoning effort=medium	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	DeepSWE v1.1 (113 long-horizon SWE tasks); GPT models run in Codex; vendor-estimated API cost/task $0.4196
DeepSWE	70.5%	HighGPT-6.1 Sol (high)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
DeepSWE	75.2%	Highreasoning effort=high	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	DeepSWE v1.1 (113 long-horizon SWE tasks); GPT models run in Codex; vendor-estimated API cost/task $0.6461
DeepSWE	73.2%	Extra highGPT-6.1 Sol (xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
DeepSWE	71.9%	Extra highreasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	DeepSWE v1.1 (113 long-horizon SWE tasks); GPT models run in Codex; vendor-estimated API cost/task $0.7886
DeepSWE	69.6%	MaxGPT-6.1 Sol (max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
DeepSWE	71.9%	Maxreasoning effort=max	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	DeepSWE v1.1 (113 long-horizon SWE tasks); GPT models run in Codex; vendor-estimated API cost/task $1.5711
FrontierCode	50.2%	Mediumgpt-6.1-sol_medium	Independent testIndependentFrontierCode (Cognition) via Epoch AI ↗	1 Oct 2026	codex	read from Epoch AI benchmark_data.zip (frontiercode_external.csv); original leaderboard https://cognition.com/frontiercode; harness codex; Mean@5
FrontierMath (Tiers 1-3)	93.7%	Maxgpt-6.1-sol_max	Independent testIndependentEpoch AI ↗	29 Sep 2026	—	FrontierMath Tiers 1-3 (v2), Epoch-run; stderr 1.4pt; data https://epoch.ai/data/benchmark_data.zip (frontiermath_tiers_1_3_v2.csv)
FrontierMath Tier 4	100.0%	Maxgpt-6.1-sol_max	Independent testIndependentEpoch AI ↗	29 Sep 2026	—	FrontierMath Tier 4 (v2), Epoch-run; stderr 0.0pt; data https://epoch.ai/data/benchmark_data.zip (frontiermath_tier_4_v2.csv)
GDP.pdf	27.0%	Lowreasoning effort=low	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	GDP.pdf (Surge AI; professional questions over complex PDFs); vendor-estimated API cost/task $0.3341
GDP.pdf	30.0%	Mediumreasoning effort=medium	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	GDP.pdf (Surge AI; professional questions over complex PDFs); vendor-estimated API cost/task $0.3375
GDP.pdf	32.0%	Highreasoning effort=high	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	GDP.pdf (Surge AI; professional questions over complex PDFs); vendor-estimated API cost/task $0.3494
GDP.pdf	31.8%	Extra highreasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	GDP.pdf (Surge AI; professional questions over complex PDFs); vendor-estimated API cost/task $0.3681
GDP.pdf	31.0%	Maxreasoning effort=max	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	GDP.pdf (Surge AI; professional questions over complex PDFs); vendor-estimated API cost/task $0.4199
GDPval-AA v2.1	1575	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	95% CI 1554.9-1595.28
GPQA Diamond	95.4%	Maxgpt-6.1-sol_max	Independent testIndependentEpoch AI ↗	29 Sep 2026	—	Epoch-run GPQA Diamond; stderr 1.4pt; data https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv)
Humanity's Last Exam	47.4%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-low)
Humanity's Last Exam	49.9%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-medium)
Humanity's Last Exam	51.4%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-high)
Humanity's Last Exam	52.6%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-xhigh)
Humanity's Last Exam	52.9%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol)
IOI (Vals)	96.9%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	—	±3.111 stderr; $1.262/test
LMArena Code Arena (WebDev)	1759	Maxgpt-6.1-sol-max	Independent testIndependentLMArena ↗	30 Sep 2026	—	Code Arena \| WebDev overall (agentic web-dev, raw); rank 3 (CI rank 3-4); 95% CI 1740-1778; 1264 votes
OSWorld 2.0 offline set	59.0%	Lowreasoning effort=low	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	OSWorld 2.0 offline set, partial reward, v2026.08.08 release; vendor-estimated API cost/task $0.4248
OSWorld 2.0 offline set	66.8%	Mediumreasoning effort=medium	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	OSWorld 2.0 offline set, partial reward, v2026.08.08 release; vendor-estimated API cost/task $0.7675
OSWorld 2.0 offline set	69.6%	Highreasoning effort=high	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	OSWorld 2.0 offline set, partial reward, v2026.08.08 release; vendor-estimated API cost/task $0.9613
OSWorld 2.0 offline set	69.4%	Extra highreasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	OSWorld 2.0 offline set, partial reward, v2026.08.08 release; vendor-estimated API cost/task $1.0466
OSWorld 2.0 offline set	71.4%	Maxreasoning effort=max	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	—	OSWorld 2.0 offline set, partial reward, v2026.08.08 release; vendor-estimated API cost/task $1.2689
SciCode	53.2%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-low)
SciCode	53.2%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-medium)
SciCode	55.8%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-high)
SciCode	55.7%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-xhigh)
SciCode	54.2%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol)
SimpleQA Verified	73.9%	Maxgpt-6.1-sol_max	Independent testIndependentEpoch AI ↗	29 Sep 2026	—	Epoch-run (no tools); ±1.39 stderr
SWE Atlas - Codebase QnA	55.1%	LowGPT-6.1 Sol (low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
SWE Atlas - Codebase QnA	60.8%	MediumGPT-6.1 Sol (medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
SWE Atlas - Codebase QnA	59.9%	HighGPT-6.1 Sol (high)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
SWE Atlas - Codebase QnA	61.0%	Extra highGPT-6.1 Sol (xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
SWE Atlas - Codebase QnA	57.8%	MaxGPT-6.1 Sol (max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component; pass@1 avg of 3 attempts
Terminal-Bench 4.0	49.0%	LowGPT-6.1 Sol (low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component run in the vendor agent harness; pass@1 avg of 3 attempts
Terminal-Bench 4.0	30.8%	LowGPT-6.1 Sol (Low)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-low)
Terminal-Bench 4.0	51.5%	MediumGPT-6.1 Sol (medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component run in the vendor agent harness; pass@1 avg of 3 attempts
Terminal-Bench 4.0	48.0%	MediumGPT-6.1 Sol (Medium)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-medium)
Terminal-Bench 4.0	51.5%	HighGPT-6.1 Sol (High)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-high)
Terminal-Bench 4.0	50.0%	HighGPT-6.1 Sol (high)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component run in the vendor agent harness; pass@1 avg of 3 attempts
Terminal-Bench 4.0	54.5%	Extra highGPT-6.1 Sol (xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component run in the vendor agent harness; pass@1 avg of 3 attempts
Terminal-Bench 4.0	54.0%	Extra highGPT-6.1 Sol (Xhigh)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol-xhigh)
Terminal-Bench 4.0	56.1%	MaxGPT-6.1 Sol (Max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	—	AA-run evaluation (AA slug gpt-6-1-sol)
Terminal-Bench 4.0	55.0%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	mini-SWE-agent	±1.821 stderr; $1.725/test
Terminal-Bench 4.0	53.0%	MaxGPT-6.1 Sol (max)	Independent testIndependentArtificial Analysis ↗	1 Oct 2026	Codex	AA Coding Agent Index component run in the vendor agent harness; pass@1 avg of 3 attempts
Terminal-Bench-Science 0.1	43.7%	Lowreasoning effort=low	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	Terminal-Bench Science 0.1; vendor-estimated API cost/task $1.7918
Terminal-Bench-Science 0.1	47.6%	Mediumreasoning effort=medium	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	Terminal-Bench Science 0.1; vendor-estimated API cost/task $2.3386
Terminal-Bench-Science 0.1	51.1%	Highreasoning effort=high	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	Terminal-Bench Science 0.1; vendor-estimated API cost/task $2.7594
Terminal-Bench-Science 0.1	53.7%	Extra highreasoning effort=xhigh	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	Terminal-Bench Science 0.1; vendor-estimated API cost/task $2.8922
Terminal-Bench-Science 0.1	57.0%	Maxreasoning effort=max	Maker's own figureVendor-reportedOpenAI ↗	29 Sep 2026	Codex	Terminal-Bench Science 0.1; vendor-estimated API cost/task $5.4652
Vals Code Migration	65.1%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	—	±4.362 stderr; $6.506/test
Vals Index	61.1%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	30 Sep 2026	—	±1.011 stderr; $3.237/test
Vals Legal Research Bench	38.5%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	—	vals id openai/gpt-6.1-sol; rank 30/72; ±3.381 stderr; $3.384561/test
Vals Public Benefits Bench v1.1	59.3%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	—	vals id openai/gpt-6.1-sol; rank 31/45; ±1.278 stderr; $0.690795/test
Vals SRE Bench	50.8%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	—	±3.095 stderr; $2.685/test
Vals Vibe Code Bench	88.9%	Maxreasoning_effort=max	Independent testIndependentVals.ai ↗	29 Sep 2026	OpenHands	±1.893 stderr; $6.235/test