Across 8 comparable benchmarks, GPT-5.1 Codex has better reported results on 6 and worse on 2.
Moderate evidence8 of 8 shared results countedHow we compare
GPT-5.1 Codex is no longer available through its official API. API prices for this version are not shown.
Price basis: Listed rates vs standard rates. The price section below shows the applicable periods.
| Fact | GPT-5.1 Codex | Haiku 4.5 |
|---|---|---|
| Context window | 400K tokens | 200K tokens |
| Input / 1M | — | $1.00 |
| Output / 1M | — | $5.00 |
| Blended (4:1) | — | $1.15 |
| Status | Retired | Available |
| Released | November 14, 2025 | October 15, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 6 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
GPT-5.1 Codex ahead on 4 of 5 counted
Haiku 4.5 ahead on 1 of 1 counted
GPT-5.1 Codex ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Science & health, Computer & tool use, Research & search, Long context, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show GPT-5.1 Codex relative to Haiku 4.5. Reading the results
| Benchmark | GPT-5.1 Codex | Haiku 4.5 | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 85.5Thinking High · tools | 62.0≠Extended · no tools | 1.38×#24 vs #69 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 70.4Thinking High · tools | 73.3≠Thinking · tools · 128K | −2.9 pts#60 vs #47 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 34.8Thinking High · tools | 27.3≠Normal · tools | +7.5 pts#50 vs #74 of 157 |
| Terminal-Bench 1.0Agentic Development · Accuracy | 56.3Thinking High · tools | 41.8≠Thinking · tools · 32K | +14.5 pts#1 vs #8 of 26 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 13.1Thinking High · tools | 11.4≠Thinking · tools | +1.7 pts#99 vs #102 of 121 |
| AIME2025Mathematics · Accuracy | 95.7Thinking High · no tools | 96.3≠Thinking · tools · 128K | −0.6 pts#22 vs #18 of 86 |
| MMMU-ProVisual Understanding · Accuracy | 72.5Thinking High · no tools | 58.6≠Thinking · no tools | +13.9 pts#80 vs #109 of 136 |
| LiveBenchCross-capability Suites · Accuracy | 68.6Normal · no tools | 61.3≠Thinking · no tools · 64K | +7.3 pts#63 vs #82 of 106 |
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Agentic Development · Accuracy
Agentic Development · Accuracy
Code Generation & Editing · Pass rate (%)
Mathematics · Accuracy
Visual Understanding · Accuracy
Cross-capability Suites · Accuracy
A dash means no result is available.
| Benchmark | GPT-5.1 Codex | Haiku 4.5 |
|---|---|---|
| CritPt | 5.7 | — |
| FrontierMath | — | 4.1 |
| FrontierMath - Tier 4 | — | 2.1 |
| GPQA Diamond | — | 73.3 |
| HLE | — | 9.7 |
| MMLU-Pro | — | 80.0 |
| SciCode | — | 42.2 |
| APEX-SWE | 29.0 | — |
| Code Migration | — | 7.5 |
| SWE-Bench Pro - Public | — | 39.5 |
| SWE-Bench Pro V2 Hard | — | 25.5 |
| Terminal Bench 2.0 | 52.8 | — |
| Claw Bench | — | 89.4 |
| MCP-Atlas | — | 40.2 |
| OSWorld 2.1 | — | 15.7 |
| Pinch Bench | — | 82.0 |
| PinchBench v2 | — | 67.7 |
| ExploitBench | — | 13.7 |
| AA Intelligence Index v5.0.1 | — | 7.4 |
| ECI | — | 142 |
| LiveBench 2.0 | — | 46.0 |
| Vals Index | — | 22.9 |
| IF Bench | — | 54.3 |
| Context Arena | — | 36.3 |
| GDP.pdf | — | 3.8 |
| MMMU | — | 73.2 |
| AA-AnalystAgent | — | 15.0 |
| AutomationBench | — | 0.46 |
| EMB (Excel Modeling) | — | 22.7 |
| Finance Agent v2 | — | 31.0 |
| Harvey Lab-AA | — | 61.0 |
| Harvey's Legal Agent Benchmark | — | 0.83 |
| Legal Research Bench | — | 10.6 |
| Public Benefits Bench v1.1 | — | 54.3 |
| SAGE | — | 31.8 |
| τ²-Bench | — | 33.0 |
| τ³-Banking | — | 9.3 |
| ARC-AGI-1 | — | 47.7 |
| ARC-AGI-2 | — | 4.5 |
| BALROG | — | 31.2 |
| Vending-Bench 2 | — | 459 |
| GAIA | — | 56.4 |
| MedCode | — | 32.7 |
| MedScribe | — | 85.2 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
GPT-5.1 Codex: Listed text rates; pricing tier not confirmed.
Haiku 4.5: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-11-13
| Price / 1M tokens | GPT-5.1 Codex | Haiku 4.5 | Ratio |
|---|---|---|---|
| Input | — | $1.00 | — |
| Output | — | $5.00 | — |
| Cached input | — | $0.10 | — |
| Cache write | — | — | — |
| Blended | — | $1.15 | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | GPT-5.1 Codex | Haiku 4.5 |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| ARC-AGI-1 | Not reported | |
| AutomationBench | Not reported | $0.07 / task Average · Pass Rate: 0.46 Reported best (effort unspecified) · zapier-private-heldout |
| LiveBench 2.0 | Not reported | $0.934 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 46.0 Max (task-specific tools) |
| Public Benefits Bench v1.1 | Not reported | $0.2245 / task Average · Criterion pass rate (%): 54.3 Unspecified Reasoning (With Tools) |
| Attribute | GPT-5.1 Codex | Haiku 4.5 |
|---|---|---|
| Availability | Retired | Available |
| Provider | OpenAI | Anthropic |
| Model type | Coding model | Multimodal model |
| First release | November 14, 2025 | October 15, 2025 |
| Context window | 400K tokens | 200K tokens |
| Max output | 128K tokens | 64K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning, Coding | Reasoning, Multilingual |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| Commercial use | Proprietary | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used GPT-5.1 Codex or Haiku 4.5? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.