Benchmark explorer.
Terminal-Bench 4.0 · AAHarness unreported4.0 · Higher is better
Source ↗Method
Artificial Analysis evaluation. Index 4.3.2 collection.
Score × cost
Where performance meets price
Terminal-Bench 4.0 · AA · 1M input + 250K output tokens · uncached USD
18 priced resultsPareto frontier
Anthropic: Claude Sonnet 5.5 ↗63.64% Terminal-Bench 4.0 · AA · $4.50 workload · Amazon Bedrock · via OpenRouter
Evaluation: Effort: max · Max, Default Fallback · Harness unreported. Price: $2.00 input / $10.00 output per 1M tokens · amazon-bedrock.
Numbers match the model key. Shared positions show an index range.
Model key Score / workload
- OpenAI: GPT-6 Luna12.63% / $0.1125
- DeepSeek: DeepSeek V4.1 Flash26.77% / $0.135
- GLM-5.3-Flash32.83% / $0.1375
- Qwen3.8-27B5.56% / $0.145
- Reflection: Beam 1.0 501B A23B3.03% / $0.50
- MiniMax-M32.02% / $0.555
- Xiaomi: MiMo-V2.6-Pro34.85% / $0.6475
- GLM-5.341.92% / $0.8763
- Step-5-Preview33.33% / $1.675
- Thinking Machines: Inkling1.01% / $1.9625
- Google: Gemini 4 Argon57.07% / $2.25
- OpenAI: GPT-6.1 Sol56.06% / $2.25
- Meta: Muse Spark Open8.08% / $2.3125
- Kimi-K312.63% / $2.529
- Anthropic: Claude Sonnet 5.563.64% / $4.50
- Anthropic: Claude Opus 5.559.6% / $9.00
- OpenAI: GPT-6 Astra59.09% / $11.25
- Anthropic: Claude Fable 5.152.02% / $22.50
Pareto frontier
- OpenAI: GPT-6 LunaScore12.63%Workload cost$0.1125OpenAI · via OpenRouter · openai/flex · tier=flex · prompt ≤ 272000 tokensEvaluation: Effort: max · Max · Harness unreported. Price: $0.05 input / $0.25 output per 1M tokens · openai/flex · tier=flex · prompt ≤ 272000 tokens.
- DeepSeek: DeepSeek V4.1 FlashScore26.77%Workload cost$0.135Decart · via OpenRouter · decart/fp4 · quantization=fp4Evaluation: Effort: max · Max · Harness unreported. Price: $0.09 input / $0.18 output per 1M tokens · decart/fp4 · quantization=fp4.
- GLM-5.3-FlashScore32.83%Workload cost$0.1375DeepInfra · via OpenRouter · deepinfra/fp4 · quantization=fp4Evaluation: Effort: max · Harness unreported. Price: $0.075 input / $0.25 output per 1M tokens · deepinfra/fp4 · quantization=fp4.
- Xiaomi: MiMo-V2.6-ProScore34.85%Workload cost$0.6475DeepInfra · via OpenRouter · deepinfra/fp8 · quantization=fp8Evaluation: Effort not specified · Harness unreported. Price: $0.43 input / $0.87 output per 1M tokens · deepinfra/fp8 · quantization=fp8.
- GLM-5.3Score41.92%Workload cost$0.8763Sail Research · via OpenRouter · sail-research/us · quantization=fp8Evaluation: Effort: max · Max · Harness unreported. Price: $0.50 input / $1.505 output per 1M tokens · sail-research/us · quantization=fp8.
- Google: Gemini 4 ArgonScore57.07%Workload cost$2.25Google AI Studio · via OpenRouter · google-ai-studio/flex · tier=flexEvaluation: Effort: high · High · Harness unreported. Price: $1.00 input / $5.00 output per 1M tokens · google-ai-studio/flex · tier=flex.
- Anthropic: Claude Sonnet 5.5Score63.64%Workload cost$4.50Amazon Bedrock · via OpenRouter · amazon-bedrockEvaluation: Effort: max · Max, Default Fallback · Harness unreported. Price: $2.00 input / $10.00 output per 1M tokens · amazon-bedrock.
Frontier: no result is cheaper with an equal or better score, or equally priced with a better score. Each point uses its lowest generally available workload price from one endpoint; evaluation and provider configurations remain visible.
| # | Model | Terminal-Bench 4.0 · AA | Configuration | API rate / unit | Source |
|---|---|---|---|---|---|
| 1 | Anthropic: Claude Sonnet 5.5anthropic | 63.64% | Effort: max · Max, Default FallbackHarness unreportedIndependent result | $2.00 in / $10.00 out/ 1M tokens · Amazon Bedrock · via OpenRouteramazon-bedrock | Result ↗Oct 18, 2026 |
| 2 | Anthropic: Claude Opus 5.5anthropic | 59.6% | Effort: max · Max, Default FallbackHarness unreportedIndependent result | $4.00 in / $20.00 out/ 1M tokens · Amazon Bedrock · via OpenRouteramazon-bedrock | Result ↗Oct 18, 2026 |
| 3 | OpenAI: GPT-6 Astraopenai | 59.09% | Effort: max · MaxHarness unreportedIndependent result | $5.00 in / $25.00 out/ 1M tokens · OpenAI · via OpenRouteropenai/flex · tier=flex · prompt ≤ 272000 tokens | Result ↗Oct 18, 2026 |
| 4 | Google: Gemini 4 Argongoogle | 57.07% | Effort: high · HighHarness unreportedIndependent result | $1.00 in / $5.00 out/ 1M tokens · Google AI Studio · via OpenRoutergoogle-ai-studio/flex · tier=flex | Result ↗Oct 18, 2026 |
| 5 | Muse Wonder FullMeta | 57.07% | Effort: max · MaxHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 6 | OpenAI: GPT-6.1 Solopenai | 56.06% | Effort: max · MaxHarness unreportedIndependent result | $1.00 in / $5.00 out/ 1M tokens · OpenAI · via OpenRouteropenai/flex · tier=flex · prompt ≤ 272000 tokens | Result ↗Oct 18, 2026 |
| 7 | Anthropic: Claude Fable 5.1anthropic | 52.02% | Effort: max · Max, Default FallbackHarness unreportedIndependent result | $10.00 in / $50.00 out/ 1M tokens · Amazon Bedrock · via OpenRouteramazon-bedrock | Result ↗Oct 18, 2026 |
| 8 | Muse WonderMeta | 52.02% | Effort: max · MaxHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 9 | GLM-5.3zai-org | 41.92% | Effort: max · MaxHarness unreportedIndependent result | $0.0632 in / $4.40 out/ 1M tokens · Relace · via OpenRouterrelace | Result ↗Oct 18, 2026 |
| 10 | Qwen3.8 Max (0902)Alibaba | 38.89% | Effort not specifiedHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 11 | Xiaomi: MiMo-V2.6-Proxiaomi | 34.85% | Effort not specifiedHarness unreportedIndependent result | $0.43 in / $0.87 out/ 1M tokens · DeepInfra · via OpenRouterdeepinfra/fp8 · quantization=fp8 | Result ↗Oct 18, 2026 |
| 12 | Step-5-Previewstepfun-ai | 33.33% | Effort not specifiedHarness unreportedIndependent result | $1.00 in / $2.70 out/ 1M tokens · DeepInfra | Result ↗Oct 18, 2026 |
| 13 | Claude Haiku 5.5Anthropic | 32.83% | Effort: max · MaxHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 14 | GLM-5.3-Flashzai-org | 32.83% | Effort: maxHarness unreportedIndependent result | $0.06 in / $1.60 out/ 1M tokens · Reka · via OpenRouterreka | Result ↗Oct 18, 2026 |
| 15 | DeepSeek: DeepSeek V4.1 Flashdeepseek | 26.77% | Effort: max · MaxHarness unreportedIndependent result | $0.09 in / $0.18 out/ 1M tokens · Decart · via OpenRouterdecart/fp4 · quantization=fp4 | Result ↗Oct 18, 2026 |
| 16 | Mistral Large 4 PreviewMistral | 26.77% | Effort not specifiedHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 17 | Grok 4.7SpaceXAI | 25.76% | Effort: xhigh · XhighHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 18 | Gemini 3.8 FlashGoogle | 19.7% | Effort: high · HighHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 19 | OpenAI: GPT-6 Lunaopenai | 12.63% | Effort: max · MaxHarness unreportedIndependent result | $0.05 in / $0.25 out/ 1M tokens · OpenAI · via OpenRouteropenai/flex · tier=flex · prompt ≤ 272000 tokens | Result ↗Oct 18, 2026 |
| 20 | Kimi-K3moonshotai | 12.63% | Effort: max · MaxHarness unreportedIndependent result | $0.27 in / $14.00 out/ 1M tokens · Relace · via OpenRouterrelace/fp4 · quantization=fp4 | Result ↗Oct 18, 2026 |
| 21 | Meta: Muse Spark Openmeta | 8.08% | Effort: xhigh · XhighHarness unreportedIndependent result | $1.25 in / $4.25 out/ 1M tokens · baseten · via Hugging Face | Result ↗Oct 18, 2026 |
| 22 | Qwen3.8-27BQwen | 5.56% | Effort: xhigh · XhighHarness unreportedIndependent result | $0.02 in / $0.50 out/ 1M tokens · Tenstorrent · via OpenRoutertenstorrent/fp4 · quantization=fp4 | Result ↗Oct 18, 2026 |
| 23 | Reflection: Beam 1.0 501B A23Breflection | 3.03% | Effort not specifiedHarness unreportedIndependent result | $0.25 in / $1.00 out/ 1M tokens · Nebius · via OpenRouternebius/fp8 · quantization=fp8 | Result ↗Oct 18, 2026 |
| 24 | MiniMax-M3MiniMaxAI | 2.02% | Effort not specifiedHarness unreportedIndependent result | $0.28 in / $1.10 out/ 1M tokens · deepinfra · via Hugging Face | Result ↗Oct 18, 2026 |
| 25 | K2 Horizon 375B A23BInstitute of Foundation Models | 1.52% | Effort not specifiedHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |
| 26 | Thinking Machines: Inklingthinkingmachines | 1.01% | Effort: xhigh · XhighHarness unreportedIndependent result | $0.95 in / $4.05 out/ 1M tokens · deepinfra · via Hugging Face | Result ↗Oct 18, 2026 |
| 27 | Nemotron 3.5 Super VLNVIDIA | 0% | Effort not specifiedHarness unreportedIndependent result | See provider quotes | Result ↗Oct 18, 2026 |