Claude Sonnet 5 vs Gemini 3.1 Pro
Choose two or three models. Copy the URL to share your selection.
| Comparison | Claude Sonnet 5claude-sonnet-5 | Gemini 3.1 Progemini-3.1-pro-preview |
|---|---|---|
| Type | Chat | Chat |
| Best for | Chat, writing, coding and reasoning tasks | Complex reasoning, multi-step analysis and difficult coding |
| TokenRex price | $2.00 in / $10.00 out per 1M tokens | $2.00 in / $12.00 out per 1M tokens |
| Cached input / 1M tokens | $0.20 per 1M tokens | $0.20 per 1M tokens |
| Cache write / 1M tokens | $2.50 per 1M tokens | No extra fee |
| Cached output | Not offered · standard output rate applies | Not offered · standard output rate applies |
| Market price | $2.00 in / $10.00 out per 1M tokens | $2.00 in / $12.00 out per 1M tokens |
| Context window (provider) | 1,000,000 tokens | Not applicable |
| Input token limit (provider) | Within the context window | 1,048,576 tokens |
| Maximum output | 128,000 tokens | 65,536 tokens |
| Knowledge cutoff | Jan 2026 | — |
| Vendor input modalities | Text, Image | Text, Image, Audio, Video |
| Vendor output modalities | Text | Text |
| TokenRex endpoints | POST /v1/chat/completions · POST /v1/messages | POST /v1/chat/completions |
| TokenRex input | Text, Image/PDF content through the chat bridge | — |
| TokenRex output | Text | — |
| Benchmarks | ||
| Benchmark evidence | 4 source-linked results | 5 source-linked results |
| GMMLU | 89% Fable 5.1 system card section 8.18 comparison table: adaptive thinking max effort, single trial, no tools or custom system prompts. Published comparison, not catalog medium effort. Anthropic Fable 5.1 system card · sections 8.1–8.11 | — |
| MILU | 89.3% Fable 5.1 system card section 8.18 comparison table: adaptive thinking max effort, single trial, no tools or custom system prompts. Published comparison, not catalog medium effort. Anthropic Fable 5.1 system card · sections 8.1–8.11 | — |
| FrontierCode v1.1 (Extended) | 49.28% Cognition FrontierCode 1.1; Claude Sonnet 5; medium effort; claude-code harness; Extended subset (150 tasks). Published weighted rubric score (new_score × 100), not the all-or-nothing pass rate; solution-source violations score zero. Source field new_score=0.4928. Publisher evaluation, not a TokenRex measurement; provider effort configuration is not a guarantee of the API's fixed thinking budget. Cognition · FrontierCode 1.1 original leaderboard | — |
| FrontierCode v1.1 (Main) | 35.2% Cognition FrontierCode 1.1; Claude Sonnet 5; medium effort; claude-code harness; Main subset (100 tasks). Published weighted rubric score (new_score × 100), not the all-or-nothing pass rate; solution-source violations score zero. Source field new_score=0.352. Publisher evaluation, not a TokenRex measurement; provider effort configuration is not a guarantee of the API's fixed thinking budget. Cognition · FrontierCode 1.1 original leaderboard | — |
| ARC-AGI-2 | — | 77.1% ARC Prize Verified, semi-private set, self-computed by Google per official gemini_3-1_pro_model_evaluation.pdf methodology (pass@1, single attempt, default gemini-3.1-pro-preview sampling). Published source text: On ARC-AGI-2 ... 3.1 Pro achieved a verified score of 77.1%. This is more than double the reasoning performance of 3 Pro. Google Blog – Gemini 3.1 Pro: A smarter model for your most complex tasks |
| GPQA Diamond | — | 94.3% Self-computed by Google per official Gemini 3.1 Pro model-evaluation methodology PDF (pass@1, single-attempt, gemini-3.1-pro-preview default settings, Feb 2026). Published source text: 94.3% on GPQA Diamond NxCode – Gemini 3.1 Pro Complete Guide 2026 (citing Google launch benchmarks) |
| SWE-bench Verified | — | 80.6% Self-computed per official methodology PDF: single-attempt scaffolding, averaged over 10x runs, gemini-3.1-pro-preview, Feb 2026. Published source text: 80.6% on SWE-Bench Verified NxCode – Gemini 3.1 Pro Complete Guide 2026 (citing Google launch benchmarks) |
| LiveCodeBench Pro Elo | — | 2887elo Public leaderboard Elo score as reported in Google's official Gemini 3.1 Pro model-evaluation PDF (Feb 2026). Published source text: LiveCodeBench Pro | 2887 Elo | Coding evaluation S5 Labs – Gemini 3.1 Pro Leads 12 of 18 Benchmarks |
| BrowseComp | — | 85.9% Deep Research configuration (search, python, browsing access) per official Gemini 3.1 Pro evaluation methodology, Feb 2026. Published source text: BrowseComp | 85.9% | Web browsing comprehension S5 Labs – Gemini 3.1 Pro Leads 12 of 18 Benchmarks |
No overall score, winner or cost-versus-score frontier is calculated without comparable, attributable benchmark evidence. Different test configurations appear separately. Vendor capabilities are distinct from TokenRex API support.
— means no published value or no matching capability. Provider token limits are not tested gateway guarantees; TokenRex media controls and response formats are listed separately.
Claude Sonnet 5
Sources and verification
Provider specifications describe the upstream model, not a guarantee of every feature through TokenRex. Prices come from the enabled catalog; benchmark results belong to the exact tested model.
- Provider model specifications
Vendor-reported
Last checked - Anthropic Fable 5.1 system card · sections 8.1–8.11
Vendor-reported
Last checked - Cognition · FrontierCode 1.1 original leaderboard
Third-party report
Last checked
Research notes (1)
- Published provider limits; maximum-capacity requests have not been independently exercised through TokenRex. Thinking-level variants use the same provider model; reasoning and answer tokens share the output budget.
Gemini 3.1 Pro
Sources and verification
Provider specifications describe the upstream model, not a guarantee of every feature through TokenRex. Prices come from the enabled catalog; benchmark results belong to the exact tested model.
- Provider model specifications
Vendor-reported
Last checked - Google Blog – Gemini 3.1 Pro: A smarter model for your most complex tasks
Vendor-reported
Last checked - NxCode – Gemini 3.1 Pro Complete Guide 2026 (citing Google launch benchmarks)
Third-party report
Last checked - S5 Labs – Gemini 3.1 Pro Leads 12 of 18 Benchmarks
Third-party report
Last checked
Research notes (1)
- Published provider limits; maximum-capacity requests have not been independently exercised through TokenRex. Thinking-level variants use the same provider model; reasoning and answer tokens share the output budget.
Popular comparisons
- GPT-6 Sol vs Claude Opus 5.5
- GPT-6 Sol vs Gemini 4 Argon
- Claude Opus 5.5 vs Gemini 4 Argon
- GPT-6.1 Sol vs GPT-6 Sol
- GPT-6 Astra vs Claude Sonnet 5
- GPT-6 Astra vs Gemini 3.1 Pro
- Claude Sonnet 5 vs Gemini 3.1 Pro
- Claude Opus 5.5 vs Claude Sonnet 5
- GPT-5.5 vs Claude Sonnet 5
- GPT-5.5 vs Gemini 2.5 Pro
- GPT-5 vs Gemini 2.5 Pro
- GPT-6 Luna vs Gemini 3.8 Flash
- Claude Haiku 4.5 vs Gemini 3.8 Flash
- GPT-5 Mini vs Gemini 2.5 Flash
- GPT-5 Nano vs Gemini 2.5 Flash Lite
- GPT-5.4 Mini vs Claude Haiku 4.5