Skip to content

OpenAI · Chat

GPT 5.4

OpenAI 🇺🇸ChatGPT familyOne of its sources says the weights aren't publicly available.The maker currently lists 2 Sep 2027 as the deprecation date. Plans can change.

Newer version: GPT 6. Our catalogue has no release date for GPT 6. We compared the version names in our catalogue.

From Netmind's listing:
GPT-5.4 is an advanced multimodal model from OpenAI with a 1M-token context window, delivering strong reasoning, high-accuracy tool calling, and improved long-horizon agentic performance across coding, analysis, and general tasks. Supports text and image inputs.
llminfo IQ
136.8
Maker's price
$5.625 blended / 1M
Max context
1.1M
Listings
98

Evidence and sources

llminfo IQ 136.8, ranked among chat models. From 489 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 112.0, 95% interval 104–120. From 28 benchmark families across 14 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedGood value (60.1). Cheaper per task than its benchmark results predict. From 59 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsTerse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. By effort: default 28 · max 56.Median 15,411 output tokens per task at default effort.How the Yap score worksReleased 5 Mar 2026Maker HQ: US (according to OpenAI's own website)
All rankings: #126 on OpenRouter this month #137 on OpenRouter this week #178 on OpenRouter today #264 on NanoGPT recentlySources: OpenAI API data, LMArena data, LiteLLM data, Models.dev data, AI/ML API data, NetMind.AI, llminfo researchPrimary data from OpenAI API from LMArena: name, variant from LiteLLM: deprecation date from Models.dev: weights availability from AI/ML API: primary mode from NetMind.AI: description from llminfo research: release date
Tags from its sources:instructtoolsvisionreasoningdocumentlong-contextfrontieragenticcodingcomputer-use

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 8 maker's prices; 2 market median prices.

Input & output

as stated by the model's providers
Text
Images
Files
Text
Images

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingPDF inputWeb searchBatch API

Listings

 OpenAIgpt-5.4Sources: GenAI Prices data, LiteLLM data, Models.dev data, OpenAI API data1.1M$2.50cache read $0.25batch $1.25$15.00batch $7.50———
 Unoroutergpt-5.4:free · freeSource: Models.dev data1.1M$0.00$0.00———
 Xpersonagpt-5.4Sources: Models.dev data, Xpersona data1.1M$0.75cache read $0.075$6.00$0.001——
 OpenRouterInference by OpenAIopenai/gpt-5.4 · flexSource: OpenRouter data docs922K$1.25cache read $0.125$7.50—65.5720.5ms
 Vercelopenai/gpt-5.4-batch · batchSource: Vercel AI Gateway data1.1M$1.25cache read $0.125$7.50———
 Tokensmindgpt-5.4Source: TokensMindNot published by the provider$1.75$10.50———
 Meganovaopenai/gpt-5.4Source: MegaNova data400K$2.00$12.00———
 Poeopenai/gpt-5.4Source: Models.dev data1.1M$2.2727cache read $0.2273$13.636———
 Amazon Web ServicesRegions: globalglobal.openai.gpt-5.4Source: LiteLLM data922K$2.50cache read $0.25$15.00———
 302aigpt-5.4Sources: 302.ai data, Models.dev data1M$2.50cache read $0.25$15.00———

15 listings without a price are hidden.

Included in plans

Copilot EnterprisegithubSubscription$39/moextra use billed past the allowancePlan page
Copilot ProgithubSubscription$10/moextra use billed past the allowancePlan page
Copilot Pro+githubSubscription$39/moextra use billed past the allowancePlan page
Copilot BusinessgithubSubscription$19/moextra use billed past the allowancePlan page
Copilot MaxgithubSubscription$100/moextra use billed past the allowancePlan page

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO67 related benchmarks · 128 results
Source: LMArena data · Nov 2026
Arena ELO
Chinese
Creative Writing
Document
Document Style-Controlled
English
Show 118 more resultsHide 118 results
Exclude Ties
Expert
French
German
Hard Prompts
Hard Prompts English
Industry Mathematical
Industry Medicine and Healthcare
Industry Software and It Services
Japanese
Korean
Longer Query
Multi Turn
Non English
Polish
Russian
Search Factuality
Spanish
Style-Controlled
Text Factuality
Text Factuality - Chinese
Text Factuality - Creative Writing
Text Factuality - English
Text Factuality - Exclude Ties
Text Factuality - Expert
Text Factuality - French
Text Factuality - German
Text Factuality - Hard Prompts
Text Factuality - Hard Prompts English
Text Factuality - Industry Mathematical
Text Factuality - Japanese
Text Factuality - Korean
Text Factuality - Longer Query
Text Factuality - Multi Turn
Text Factuality - Non English
Text Factuality - Polish
Text Factuality - Russian
Text Factuality - Spanish
Vision
Vision - Chinese
Vision - Creative Writing Vision
Vision - Diagram
Vision - English
Vision - Entity Recognition
Vision - Homework
Vision - Humor
Vision Style-Controlled
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Reasoning

ARC-AGI4 related benchmarks · 12 results
ARC-AGI
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
ARC-AGI 2
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
ARC-AGI-1
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
v2
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
CL-bench9 related benchmarks · 34 results
Overall
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
benchmark-owner leaderboard run · site model-mode label: GPT-5.4 (xhigh) · 1899 tasks · GPT-5.1 judge · low reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
Domain Knowledge Reasoning
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
benchmark-owner leaderboard run · site model-mode label: GPT-5.4 (xhigh) · 1899 tasks · GPT-5.1 judge · low reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
Empirical Discovery & Simulation
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
benchmark-owner leaderboard run · site model-mode label: GPT-5.4 (xhigh) · 1899 tasks · GPT-5.1 judge · low reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
Life · Behavioral Records & Activity Trails
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
By site model-mode label · benchmark-owner leaderboard run · 405 tasks · GPT-5.1 judge · high reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
benchmark-owner leaderboard run · site model-mode label: GPT 5.4 (High) · 405 tasks · GPT-5.1 judge · high reasoning effort · candidate API snapshot and exact candidate harness not specified on public leaderboard
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Life · Communication & Social Interactions
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
By site model-mode label · benchmark-owner leaderboard run · 405 tasks · GPT-5.1 judge · high reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
benchmark-owner leaderboard run · site model-mode label: GPT 5.4 (High) · 405 tasks · GPT-5.1 judge · high reasoning effort · candidate API snapshot and exact candidate harness not specified on public leaderboard
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Life · Fragmented Information & Revisions
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
By site model-mode label · benchmark-owner leaderboard run · 405 tasks · GPT-5.1 judge · high reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
benchmark-owner leaderboard run · site model-mode label: GPT 5.4 (High) · 405 tasks · GPT-5.1 judge · high reasoning effort · candidate API snapshot and exact candidate harness not specified on public leaderboard
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Show 10 more resultsHide 10 results
Life · Overall
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
By site model-mode label · benchmark-owner leaderboard run · 405 tasks · GPT-5.1 judge · high reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
benchmark-owner leaderboard run · site model-mode label: GPT 5.4 (High) · 405 tasks · GPT-5.1 judge · high reasoning effort · candidate API snapshot and exact candidate harness not specified on public leaderboard
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Effort not reported
Procedural Task Execution
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
benchmark-owner leaderboard run · site model-mode label: GPT-5.4 (xhigh) · 1899 tasks · GPT-5.1 judge · low reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
Rule System Application
Reported by Tencent Hunyuan / Fudan NLP Group · via BenchBoard docs
benchmark-owner leaderboard run · site model-mode label: GPT-5.4 (xhigh) · 1899 tasks · GPT-5.1 judge · low reasoning effort · single run; no standard deviation · candidate API snapshot and exact candidate harness not specified on public leaderboard
GraphwalksSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
OfficeQAOpenAI GPT-5.4 launch table · Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M2.7 official benchmark overview · exact harness/version not stated
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table

Agentic

BrowseComp2 results
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026
FinanceAgent v1.1OpenAI GPT-5.4 launch table · Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GAIA4 results
Source: GAIA Leaderboard data · Apr 2026
Difficulty level
Difficulty not reported
GDPvalReported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GDPval-AA · Published ScoreMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MCP Atlas2 results
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
MLE-Bench LiteMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Online-Mind2Web8 related benchmarks · 9 results
Online-Mind2Web
Source: Online-Mind2Web Leaderboard data · Aug 2026
Setting not stated
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
screenshot observations only
Easy
Source: Online-Mind2Web Leaderboard data · Aug 2026
Show 2 more resultsHide 2 results
OpenHandsSource: OpenHands Index data
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
SWE-Bench Pro3 results
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M2.7 official benchmark overview · exact harness/version not stated
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Source: OpenRouter benchmark exports · Oct 2026
5 Mar 2026 version · 90-day results
Across 3 providers lowest 72.3% median 72.7% highest 73.1%OpenRouter's automatic routing 75%
Show results from 3 providers and OpenRouter's automatic routingHide results from 3 providers and OpenRouter's automatic routing
  • OpenRouter's automatic routing
  • Azure
  • OpenAI
  • Amazon Bedrock
WebArena-VerifiedDOM + screenshot interaction · Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Coding

Arena ELO7 related benchmarks · 17 results
Source: LMArena data · Nov 2026
Coding
Text Factuality - Coding
Webdev
Webdev - Image to Webdev
Webdev - Webdev
Webdev - Webdev Html
Show 3 more resultsHide 3 results
Webdev - Webdev React
Multi-SWE-BenchMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
ValsSource: Vals AI

Long context

GDM-MRCRv2: 8 needle8 related benchmarks · 40 results
Source: Context Arena data · Apr 2026
Benchmarklowmediumhighxhighnot stated
GDM-MRCRv2: 8 needle
8k
16k
32k
Show 4 more benchmarksHide 4 benchmarks
Benchmarklowmediumhighxhighnot stated
64k
128k
256k
512k
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Show 2 more resultsHide 2 results

Vision

MMMU-Pro3 results
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

General

Math

FrontierMath5 related benchmarks · 7 results
FrontierMath
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Tier 1-3 (v2)
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
Tier 4 (v2)
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
MathVisionSenseNova 6.7 Flash-Lite official benchmark chart · fixed boxed-answer prompt · peers use the higher reported boxed/unboxed run · Reported by SenseTime / OpenSenseNova · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

OCR

OmniDocBenchSource: Nanonets IDP data

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond10 results
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI GPT-5.4 launch table
Source: OpenRouter benchmark exports · Oct 2026
5 Mar 2026 version · 90-day results
Across 3 providers lowest 90% median 90.2% highest 90.3%OpenRouter's automatic routing 90.2%
Show results from 3 providers and OpenRouter's automatic routingHide results from 3 providers and OpenRouter's automatic routing
  • OpenAI
  • OpenRouter's automatic routing
  • Amazon Bedrock
  • Azure

Multilingual

EKI EstonianSource: Keelemudelite mõõdupuu data docs

Instruction following

Knowledge

Safety

Factuality

Speech

Other

BigLaw BenchReported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Claw-Eval v1.1Reported by Claw-Eval · via BenchBoard docs
GraphWalks BFSF1 · Reported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
HealthBench Hardlength-adjusted, max reasoning effort, GPT-5.6 system card Table 6 column GPT-5.4 (30.3 unadjusted, 2161 chars) · Source: HealthBench Hard (Arcophos) data docs · Reported by the model's developer. The source relaying it did not verify it. · Jun 2026
Horangi36 related benchmarks · 36 results
MM-ClawBenchMiniMax M2.7 official benchmark overview · exact harness/version not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Nejumi 428 related benchmarks · 56 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
Show 44 more resultsHide 44 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_表現
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
By openai/gpt-5.4-2026-03-05
NovaPPTBenchSenseNova 6.7 Flash-Lite official benchmark chart · internal benchmark · provider states it will be open-sourced · Reported by SenseTime / OpenSenseNova · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
RNG-Bench12 related benchmarks · 12 results
Duel Elo
Reported by RNG-Bench / InternLM · via BenchBoard docs
Matching Pairs · image · poker theme · 16 games per model
3D Maze Efficiency
Reported by RNG-Bench / InternLM · via BenchBoard docs
successful episodes only · 13×13 · no minimap · mean optimal path 60 steps
3D Maze Explore Rate
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 13×13 · no minimap · mean optimal path 60 steps
3D Maze Game Score
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 13×13 · no minimap · mean optimal path 60 steps
3D Maze Success Rate
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 13×13 · no minimap · mean optimal path 60 steps
3D Maze Wall Collisions
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 13×13 · no minimap · mean optimal path 60 steps
Show 6 more resultsHide 6 results
Duel Score
Reported by RNG-Bench / InternLM · via BenchBoard docs
Matching Pairs · image · poker theme · 16 games per model
Duel Win Rate
Reported by RNG-Bench / InternLM · via BenchBoard docs
Matching Pairs · image · poker theme · 16 games per model
Matching Pairs Invalid Action
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 10×10 · image · noise theme
Matching Pairs Parse Failure
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 10×10 · image · noise theme
Matching Pairs Responses per Score
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 10×10 · image · noise theme
Matching Pairs Score
Reported by RNG-Bench / InternLM · via BenchBoard docs
single-player · 10×10 · image · noise theme
Sayf16 related benchmarks · 16 results
Show 10 more resultsHide 10 results
SkillsBench7 related benchmarks · 7 results
Vals · With Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
Copilot resolution-rate difference
Reported by GitHub · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GitHub 2026-06-20 weekly suite · Copilot CLI vs Codex CLI with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated
Vals · With Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
Vals · With Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · With Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
Vals · Without Skills · Accuracy
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
Vals · Without Skills · Mean cost / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
Show 1 more resultHide 1 result
Vals · Without Skills · Mean duration / test
Reported by Vals AI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Vals AI · updated 2026-09-27 · metadata version 1 · 87 tasks in trace charts · 3 trials/task/condition · per-task mean then task average · OpenHands-CLI@ · Without Skills · system=openai/gpt-5.4-2026-03-05 · provider=OpenAI · reasoning_effort=xhigh · max_output_tokens=128000
τ²-bench TelecomReported by OpenAI · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · LMArena, CC BY 4.0 (licence) · BullshitBench by Peter Gostev, MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · Steel.dev Agent Leaderboards, MIT (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · CyberGym leaderboard (Wang, Shi, He, Cai, Zhang, Song; UC Berkeley), ICLR 2026. Results evaluated and submitted by the listed teams. No licence is stated for the result data; the repository's Apache-2.0 licence covers its code only. · Keelemudelite mõõdupuu: Independent LLM leaderboard for Estonian. Eesti Keele Instituut. https://moodupuu.eki.ee, CC BY 4.0. Latest per-benchmark results selected with partial-run flags retained. (licence) · ExploitGym leaderboard (Wang et al., UC Berkeley), arXiv:2605.11086. No licence is stated for the result data; the repository's Apache-2.0 licence covers its code only, and task data keeps its upstream licences. Literal exploit counts; hidden rows omitted. · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Arcophos. HealthBench Hard Leaderboard. https://healthbenchhard.ai. CC BY 4.0. Source document linked on each result. (licence) · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MineBench, MIT (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Online-Mind2Web Leaderboard (OSU NLP), Apache-2.0 (licence) · OpenHands Index (All Hands AI), Apache-2.0 (licence) · Sayf-eval. Copyright (c) 2026 Qatar Computing Research Institute (QCRI). MIT License. (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · Swallow LLM Leaderboard v2, Swallow LLM team (Institute of Science Tokyo). Contents CC BY 4.0; Japanese post-trained task scores on the source's 0..1 scale. (licence) · Sierra Research tau2-bench leaderboard submissions (github.com/sierra-research/tau2-bench), MIT (licence) · Vectara Hallucination Leaderboard, Apache-2.0 (licence)

Around the web

Also known as: aihubmix/gpt-5.4-high, aihubmix/gpt-5.4-low, azure/eu/gpt-5.4, azure/eu/gpt-5.4-2026-03-05, azure/gpt-5.4, azure/gpt-5.4-2026-03-05, azure/gpt-5.4@eastus2, azure/gpt-5.4@francecentral, azure/gpt-5.4@swedencentral, azure/us/gpt-5.4, azure/us/gpt-5.4-2026-03-05, azure_ai/gpt-5.4, azure_ai/gpt-5.4-2026-03-05, bedrock/gpt-5.4@us-east-1, bedrock/gpt-5.4@us-east-2, bedrock/gpt-5.4@us-west-2, bedrock_mantle/openai.gpt-5.4, bedrock_mantle/us-gov-east-1/openai.gpt-5.4, bedrock_mantle/us-gov-west-1/openai.gpt-5.4, chatgpt/gpt-5.4, databricks-gpt-5-4, databricks/databricks-gpt-5-4, github_copilot/gpt-5.4, global.openai.gpt-5.4, gmi/openai/gpt-5.4, gpt-5-4, gpt-5.4-2026-03-05, gpt-5.4-high, gpt-5.4-high-codex-harness, gpt-5.4-low, gpt-5.4-medium-codex-harness, gpt-5.4-search, gpt-5.4:free, gpt-5.4@eu, gpt-54, openai-gpt-5.4, openai-gpt-54, openai.gpt-5.4, openai/gpt-5-4, openai/gpt-5.4, openai/gpt-5.4-2026-03-05, openai/gpt-5.4-batch, openai/gpt-5.4-fast, openai/gpt-5.4:batch, perplexity/openai/gpt-5.4, replicate/openai/gpt-5.4, us-gov.openai.gpt-5.4, us.openai.gpt-5.4

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works