BenchLM comparison
ChatGPT vs Claude (July 2026): Benchmarks, Price & Verdict
Claude leads the public ranking comparison: Claude Fable 5 scores 83.68 at #2, versus GPT-5.6 Sol at 81.96 and #3. ChatGPT wins on price: GPT-5.6 Sol costs less per API token.
This page compares ChatGPT and Claude as products, not just as models. Concretely, that means the flagship models behind each app: ChatGPT currently runs on OpenAI's GPT-5.6 family — Sol, Terra, and Luna(launched July 9, 2026), while the Claude app runs on Anthropic's Claude Fable 5, Claude Opus 4.8, and Claude Sonnet 5. Every number below comes from BenchLM's sourced benchmark and pricing data for those specific models.
Whether you search for "ChatGPT vs Claude" or "Claude vs ChatGPT", the question is the same: which assistant should get your $20 a month, and which API should run your workload? Here is what the data says in July 2026.
The short answer
Claude has the stronger public score of the two products. Claude Fable 5 posts 83.68 at #2, while GPT-5.6 Sol posts 81.96 at #3. Fable 5 also records 95 on SWE-bench Verified and 84.3 on Terminal-Bench 2.
ChatGPT's answer is the GPT-5.6 family. GPT-5.6 Sol is publicly ranked with an Estimated evidence label, and its sourced category results include 91.9 on Terminal-Bench 2 and 92.2 on BrowseComp. The table below keeps that public score separate from the raw benchmark rows.
On price the roles flip. GPT-5.6 Sol's API rate is $5 input / $30 output per million tokens against $10 / $50 for Claude Fable 5, and ChatGPT Plus at $20/month remains the default consumer subscription. If you want the highest benchmarked ceiling today, it is Claude; if you want more capability per dollar and the broader consumer feature set, it is ChatGPT.
ChatGPT vs Claude: the models behind the apps
A product comparison only makes sense at the model level, so here are the current flagships side by side. Public scores and ranks come from the same lane as the main leaderboard; Arena Elo is shown separately where available.
| Model | BenchLM score | Arena Elo | Context |
|---|---|---|---|
| GPT-5.6 SolOpenAI | 81.96 | 1486.4 | 1M |
| GPT-5.6 TerraOpenAI | 72.57 | — | 1M |
| GPT-5.6 LunaOpenAI | 67.17 | — | 1M |
| GPT-5.5OpenAI | 73.51 | 1481.0 | 1M |
| Claude Fable 5Anthropic | 83.68 | 1506.8 | 1M+ |
| Claude Opus 4.8Anthropic | 78.34 | 1484.6 | 1M |
| Claude Sonnet 5Anthropic | 65.32 | 1461.4 | 1M |
Want the benchmark-by-benchmark breakdown? Every pairing has a dedicated model comparison page — start with Claude Fable 5 vs GPT-5.6 Sol for the flagship matchup, or Claude Opus 4.8 vs GPT-5.6 Sol if you use Opus for daily work.
ChatGPT vs Claude pricing: subscriptions and API
On subscriptions, ChatGPT Plus costs $20/monthand unlocks the GPT-5.6 models with higher usage limits. Anthropic's Claude Pro sits in the same consumer price band with usage caps that scale up on the more expensive Max tiers — Claude's subscription limits are generally tighter than ChatGPT's at the entry level, reflecting how expensive Fable 5 is to serve.
The API rates are where the gap is measurable. Sol undercuts Fable 5 by half on both input and output, Terra matches Claude Sonnet 5's standard rate, and Luna is the cheapest model on this page. Full provider rate cards live on the OpenAI API pricing and Anthropic API pricing hubs.
| Model | Input / 1M tokens | Output / 1M tokens | Note |
|---|---|---|---|
| GPT-5.6 Sol | $5.00 | $30.00 | — |
| GPT-5.6 Terra | $2.50 | $15.00 | — |
| GPT-5.6 Luna | $1.00 | $6.00 | — |
| GPT-5.5 | $5.00 | $30.00 | Previous ChatGPT flagship, still on the API |
| Claude Fable 5 | $10.00 | $50.00 | — |
| Claude Opus 4.8 | $5.00 | $25.00 | — |
| Claude Sonnet 5 | $2.00 | $10.00 | Introductory rate through Aug 31, 2026; $3 / $15 standard |
Which should you pick?
Pick ChatGPT if…
- You want more capability per dollar: GPT-5.6 Sol is $5 / $30 per million tokens versus $10 / $50 for Claude Fable 5, and Luna drops to $1 / $6.
- Agentic and terminal work matters: Sol's early results lead this matchup on Terminal-Bench 2 (91.9 vs 84.3) and BrowseComp (92.2 vs 86.9).
- You want one $20/month app that covers the widest consumer surface — voice, images, search, and custom GPTs alongside the core models.
- You would rather bet on the newest release: GPT-5.6 is the freshest flagship family here, with benchmark coverage still filling in.
- You need a cheap high-volume tier — GPT-5.6 Luna at $1 / $6 has no Claude equivalent at that price.
Pick Claude if…
- You want the higher public score: Claude Fable 5 posts 83.68, versus 81.96 for GPT-5.6 Sol.
- Coding is the job: Fable 5's 95 on SWE-bench Verified and 80 on SWE-bench Pro lead this comparison (Sol posts 64.6 on SWE-bench Pro).
- Writing quality and conversational judgment matter — Fable 5's 1508.5 Arena Elo is the top human-preference score BenchLM tracks.
- You work with huge documents: Fable 5 is the only model here rated at a 1M+ context window.
- You want a strong mid-tier deal: Claude Sonnet 5 is $2 / $10 on introductory pricing through August 31, 2026.
How this comparison works
BenchLM compares products through the models that power them, because "ChatGPT" and "Claude" are moving targets — each app swaps in new flagships several times a year. Overall scores use the current public ranking lane; Arena Elo is the community human-preference rating where one has been published; API prices come from each provider's official rate card as recorded in BenchLM's pricing data. Models outside the public ranking show as unranked instead of receiving a legacy fallback. Individual benchmark scores quoted in the text (SWE-bench Verified, Terminal-Bench 2, BrowseComp) come from the same sourced dataset that powers every model comparison page.
ChatGPT vs Claude FAQ
Is Claude better than ChatGPT?
On the current public ranking, Claude leads: Claude Fable 5 scores 83.68 at #2, while GPT-5.6 Sol scores 81.96 at #3. Sol carries an Estimated evidence label, so the point gap should not be treated as a guarantee for every workload.
Is Claude cheaper than ChatGPT?
No — at the flagship level Claude is roughly twice the price. Claude Fable 5 costs $10 input / $50 output per million tokens on the API versus $5 / $30 for GPT-5.6 Sol, and nothing in Claude's lineup matches GPT-5.6 Luna's $1 / $6. The exception is the mid-tier: Claude Sonnet 5's $2 / $10 introductory rate undercuts GPT-5.6 Terra's $2.50 / $15 until August 31, 2026.
Which is better for coding, ChatGPT or Claude?
Claude, on the sourced numbers. Claude Fable 5 posts 95 on SWE-bench Verified and 80 on SWE-bench Pro — the best coding line in this comparison — with Claude Opus 4.8 close behind at 88.6 SWE-bench Verified. GPT-5.6 Sol counters on agentic terminal work, where its 91.9 Terminal-Bench 2 beats Fable 5's 84.3, so heavy CLI-agent users may still prefer ChatGPT.
Can I use both ChatGPT and Claude?
Yes, and many people do — two entry subscriptions cost about as much as one premium tier. A common split is Claude for coding and long-form writing, where Fable 5's SWE-bench and Arena Elo numbers lead, and ChatGPT for everyday questions, browsing-style research, and high-volume tasks on the cheaper GPT-5.6 tiers. On the API side, mixing providers per task is standard practice.