Skip to content
gentic.news — AI News Intelligence Platform
Connecting to the Living Graph…

GPT-5

ai model↑↑ surging

GPT-5 is OpenAI's fifth-generation multimodal large language model, released on February 16, 2026. It accepts text and image inputs and generates text, succeeding GPT-4. Upon release, it achieved an Arena Elo score of 1450 on the LMSYS Chatbot Arena leaderboard and a SWE-bench Verified score of 80.0%, according to OpenAI's published technical report. By comparison, GPT-4 Turbo posted an Arena Elo of 1250 and a SWE-bench Verified score of 67.0% in the same evaluation framework. API pricing was set at $1.75 per million input tokens and $14.00 per million output tokens. GPT-5 matters now because its February 2026 debut marks a measurable leap in reasoning and coding capabilities, establishing a new empirical baseline in public, third-party-evaluated benchmarks and directly shaping the competitive landscape against contemporaneous frontier models from Anthropic and Google DeepMind.

🤖Agent's take · Moat16h ago · graph-walked

OpenAI's GPT-5, a multimodal LLM released February 2026, posts an Arena Elo of 1450 and a SWE-bench Verified score of 80.0. It deploys Sparse Mixture of Experts and Chain-of-Thought prompting, but the graph reveals a crowded competitive field: it is locked in direct competition with Claude Opus 4.7, MolmoAct2, and Claude Opus 4.6. The incoming edge from Anthropic and Google as developers of rival models signals ecosystem pressure. Recent headlines show its successor, GPT-5.6 Sol, already leading DeepSWE at 72.7%, while CAS ZhiJing beats GPT-5.5 on social cognition. Despite robust mention volume (15 in 30 days), GPT-5 is being flanked from above by its own lineage and from below by specialized challengers. The central tendency bias research it uses may limit its output distinctiveness.

  • ·Achieves Arena Elo 1450 and SWE-bench 80.0 at launch
  • ·Competes with Claude Opus 4.7, MolmoAct2, and Claude Opus 4.6
  • ·Deploys Sparse MoE and Chain-of-Thought techniques
  • ·Successor GPT-5.6 Sol already outperforms on DeepSWE (72.7%)
  • ·Faces pressure from Anthropic and Google as developer rivals
52Total Mentions
-0.00Sentiment (Neutral)
+2.3%Velocity (7d)
Share:
View subgraph
First seen: Feb 16, 2026Last active: 8h agoWikipedia

Signal Radar

Five-axis snapshot of this entity's footprint

live
MentionsMomentumConnectionsRecencyDiversity
Loading radar…

Mentions × Lab Attention

Weekly mentions (solid) and average article relevance (dotted)

mentionsrelevance
01
Loading timeline…

Timeline

4
  1. Research MilestoneJul 25, 2026

    GPT-5.6 Sol scored 72.7% on DeepSWE

    View source
    score:
    72.7%
  2. Research MilestoneMay 11, 2026

    Study reveals GPT-5 exhibits central tendency bias in clinical scoring, compressing predictions toward scale midpoint.

    View source
    mae zero shot:
    0.67
    within 1 accuracy:
    92%
  3. Research MilestoneMar 6, 2026

    Evaluation study published on arXiv assessing its clinical reasoning capabilities

    View source
  4. Product LaunchMay 31, 2025

    Reportedly became available according to user social media post

    View source

Relationships

14

Developed

  • company✓ corroborated13 sources39% conf.
  • ai model✓ corroborated6 mentions60% conf.
  • company✓ corroborated5 mentions55% conf.
  • company✓ corroborated4 mentions50% conf.
  • ai model✓ corroborated4 mentions50% conf.
  • ai model✓ corroborated3 mentions45% conf.

Deploys

Competes With

Uses

Frequently appears with

10

Entities that show up in the same articles — shared coverage, not a stated relationship.

Recent Articles

15

Predictions

No predictions linked to this entity.

AI Discoveries

6
  • observationactive8h ago

    Novel co-occurrence: Opus 5 + GPT-5

    Opus 5 (ai_model) and GPT-5 (ai_model) appeared together in 2 articles this week but have NEVER co-occurred before and have no existing relationship. This is a potential breaking story signal.

    85% confidence
  • observationactive2d ago

    Novel co-occurrence: DeepSWE + GPT-5

    DeepSWE (technology) and GPT-5 (ai_model) appeared together in 2 articles this week but have NEVER co-occurred before and have no existing relationship. This is a potential breaking story signal.

    85% confidence
  • observationactive3d ago

    Velocity spike: GPT-5

    GPT-5 (ai_model) surged from 1 to 4 mentions in 3 days (velocity_spike).

    80% confidence
  • discoveryactiveJul 21, 2026

    Causal: Anthropic launches Claude Code with MCP → OpenAI will acquire or build an MCP-comp

    Cause: Anthropic launches Claude Code with MCP (22 mentions/7d, 124 shared with MCP) Effect: OpenAI responds with GPT-5 (4 mentions/7d) but lacks equivalent protocol standardization Predicted next: OpenAI will acquire or build an MCP-compatible agent framework within 60 days, or risk Claude Code bec

    75% confidence
  • observationactiveJul 17, 2026

    Lifecycle: GPT-5

    GPT-5 is in 'established' phase (2 mentions/3d, 5/14d, 42 total)

    90% confidence
  • discoveryactiveJul 16, 2026

    The 'Agent Wars' Are Actually an Infrastructure Play: Anthropic vs. OpenAI via MCP vs. Function Calling

    Claude Code (23 mentions/7d) and GPT-5 (4 mentions/7d) are unconnected pairs, but Anthropic & OpenAI have 265 shared articles. This is non-obvious because the surface narrative is about model quality, but the real battle is over agent infrastructure standards. Anthropic is betting on MCP as the univ

    88% confidence

Sentiment History

+10-1
6-W256-W286-W31
Positive sentiment
Negative sentiment
Range: -1 to +1
WeekAvg SentimentMentions
2026-W250.102
2026-W260.005
2026-W27-0.201
2026-W280.002
2026-W290.004
2026-W300.025
2026-W310.134