Benchmarks / Artificial Analysis

Reported by Artificial Analysis

Artificial Analysis

Share of Terminal-Bench 2.1 tasks completed, as run by Artificial Analysis with its own harness and prompts.

Last updated 20 Oct 2026

Results dated
20 Oct 2026
Results
188 configurations of 131 models
Unit
% of tasks
Licence
Artificial Analysis commercial data licence

Terminal-Bench 2.1: Muse Spark Open

Top 15 of 188 results · % of tasks, higher is better. Choose a model to highlight it.Clear highlight

  1. 1 Claude Fable 5.1 (max reasoning)Anthropic 91.4%
  2. 2 Claude Fable 5.1 (extra-high reasoning)Anthropic 91.0%
  3. 3 Claude Fable 5.1 (high reasoning)Anthropic 89.9%
  4. 3 GPT-6 Astra (high reasoning)OpenAI 89.9%
  5. 5 GPT-5.6 Sol (extra-high reasoning)OpenAI 89.5%
  6. 5 GPT-6 Astra (medium reasoning)OpenAI 89.5%
  7. 7 Claude Opus 5 (max reasoning)Anthropic 89.1%
  8. 7 GPT-6 Astra (extra-high reasoning)OpenAI 89.1%
  9. 9 Qwen3.8-Max (0902)Alibaba 88.8%
  10. 10 GPT-6 Astra (max reasoning)OpenAI 88.4%
  11. 10 Grok 4.6 (high reasoning)xAI 88.4%
  12. 12 Claude Fable 5.1 (medium reasoning)Anthropic 88.0%
  13. 12 Claude Opus 5 (extra-high reasoning)Anthropic 88.0%
  14. 12 GPT-5.6 Sol (max reasoning)OpenAI 88.0%
  15. 12 GPT-5.6 Terra (max reasoning)OpenAI 88.0%
  16. 60 Muse Spark Open (extra-high reasoning)Meta 76.4%

Full results

Artificial Analysis: Terminal-Bench 2.1, % of tasks, higher is better
#ModelTerminal-Bench 2.1
% of tasks, higher is better		Price
$ per million tokens, in / out
1 Claude Fable 5.1 (max reasoning)Anthropic
91.4%
$10 / $50
2 Claude Fable 5.1 (extra-high reasoning)Anthropic
91.0%
$10 / $50
3 Claude Fable 5.1 (high reasoning)Anthropic
89.9%
$10 / $50
3 GPT-6 Astra (high reasoning)OpenAI
89.9%
$10 / $50
5 GPT-5.6 Sol (extra-high reasoning)OpenAI
89.5%
$4 / $20
5 GPT-6 Astra (medium reasoning)OpenAI
89.5%
$10 / $50
7 Claude Opus 5 (max reasoning)Anthropic
89.1%
$5 / $25
7 GPT-6 Astra (extra-high reasoning)OpenAI
89.1%
$10 / $50
9 Qwen3.8-Max (0902)Alibaba
88.8%
$2 / $6
10 GPT-6 Astra (max reasoning)OpenAI
88.4%
$10 / $50
10 Grok 4.6 (high reasoning)xAI
88.4%
$2 / $6
12 Claude Fable 5.1 (medium reasoning)Anthropic
88.0%
$10 / $50
12 Claude Opus 5 (extra-high reasoning)Anthropic
88.0%
$5 / $25
12 GPT-5.6 Sol (max reasoning)OpenAI
88.0%
$4 / $20
12 GPT-5.6 Terra (max reasoning)OpenAI
88.0%
$2 / $12
12 GPT-6 Astra (low reasoning)OpenAI
88.0%
$10 / $50
12 Grok 4.6 (extra-high reasoning)xAI
88.0%
$2 / $6
18 Claude Opus 5 (high reasoning)Anthropic
87.6%
$5 / $25
18 Gemini 3.8 Flash (high reasoning)Google
87.6%
$1.50 / $7.50
20 GPT-5.6 Sol (high reasoning)OpenAI
87.3%
$4 / $20
21 Qwen3.8-Flash-NextAlibaba
86.1%
–
21 Claude Opus 5 (medium reasoning)Anthropic
86.1%
$5 / $25
21 GPT-5.6 Sol (medium reasoning)OpenAI
86.1%
$4 / $20
24 Gemini 3.7 Flash (high reasoning)Google
85.8%
$1.50 / $7.50
25 Muse Spark 1.3 (extra-high reasoning)Meta
85.4%
$1.25 / $4.25
26 Claude Fable 5.1 (low reasoning)Anthropic
85.0%
$10 / $50
26 Kimi K3 (max reasoning)Moonshot AI
85.0%
$3 / $15
28 Claude Fable 5 (max reasoning)Anthropic
84.6%
$10 / $50
28 Claude Opus 4.8 (max reasoning)Anthropic
84.6%
$5 / $25
30 Muse Spark 1.3 (max reasoning)Meta
84.3%
$1.25 / $4.25
30 GPT-5.5 (extra-high reasoning)OpenAI
84.3%
$5 / $30
30 Grok 4.6 (medium reasoning)xAI
84.3%
$2 / $6
30 GLM-5.3-FlashZ.ai
84.3%
$0.15 / $0.50
34 Gemini 3.8 Flash (medium reasoning)Google
83.9%
$1.50 / $7.50
34 GLM-5.3 (max reasoning)Z.ai
83.9%
$1.40 / $4.40
36 Claude Opus 4.7 (max reasoning)Anthropic
83.1%
$5 / $25
36 Gemini 3.8 Flash (low reasoning)Google
83.1%
$1.50 / $7.50
38 Kimi K3 (low reasoning)Moonshot AI
82.4%
$3 / $15
39 Qwen3.8-2.4T-A95BAlibaba
82.0%
$2 / $6
40 Grok 4.5 (high reasoning)xAI
81.6%
$2 / $6
41 Qwen3.8-Max (0803)Alibaba
81.3%
–
42 GPT-5.6 Luna (max reasoning)OpenAI
80.9%
$0.20 / $1.20
43 Claude Sonnet 5 (max reasoning)Anthropic
80.5%
$2 / $10
43 GPT-5.5 (medium reasoning)OpenAI
80.5%
$5 / $30
45 Muse Spark 1.2 (extra-high reasoning)Meta
80.2%
$1.25 / $4.25
45 GPT-5.6 Terra (extra-high reasoning)OpenAI
80.2%
$2 / $12
47 Qwen3.8-27B (extra-high reasoning)Alibaba
79.8%
$0.50 / $3
47 Gemini 3.7 Flash (low reasoning)Google
79.8%
$1.50 / $7.50
49 GPT-5.5 (high reasoning)OpenAI
79.4%
$5 / $30
50 DeepSeek-V4-Flash (0731, max reasoning)DeepSeek
78.7%
$0.14 / $0.28
50 DeepSeek-V4-Pro (0813, max reasoning)DeepSeek
78.7%
$1.32 / $3.96
50 Gemini 3.5 Flash (high reasoning)Google
78.7%
$1.50 / $9
53 Gemini 3.7 Flash (medium reasoning)Google
78.3%
$1.50 / $7.50
53 GPT-5.4 (extra-high reasoning)OpenAI
78.3%
$2.50 / $15
55 Muse Spark 1.1 (extra-high reasoning)Meta
77.9%
$1.25 / $4.25
55 GPT-5.6 Luna (extra-high reasoning)OpenAI
77.9%
$0.20 / $1.20
55 GLM-5.2 (max reasoning)Z.ai
77.9%
$1.40 / $4.40
58 Gemini 3.6 Flash (high reasoning)Google
77.5%
$1.50 / $7.50
59 GPT-5.6 Sol (low reasoning)OpenAI
76.8%
$4 / $20
60 Claude Opus 5 (low reasoning)Anthropic
76.4%
$5 / $25
60 Muse Spark Open (extra-high reasoning)Meta
76.4%
–
62 GPT-5.6 Terra (high reasoning)OpenAI
75.7%
$2 / $12
63 Claude Sonnet 5 (no reasoning)Anthropic
75.3%
$2 / $10
63 Grok 4.6 (low reasoning)xAI
75.3%
$2 / $6
65 Qwen3.7-MaxAlibaba
74.5%
$1.48 / $4.42
66 DeepSeek-V4-Flash-Vision-Exp (max reasoning)DeepSeek
74.2%
$0.44 / $1.32
66 GPT-5.6 Sol (no reasoning)OpenAI
74.2%
$4 / $20
68 Gemini 3.1 Pro PreviewGoogle
73.8%
$2 / $12
69 GPT-5.6 Terra (medium reasoning)OpenAI
72.3%
$2 / $12
70 Claude Sonnet 4.6 (max reasoning)Anthropic
71.2%
$3 / $15
71 GPT-5.6 Luna (high reasoning)OpenAI
69.7%
$0.20 / $1.20
72 Qwen3.8-27B (low reasoning)Alibaba
67.4%
$0.50 / $3
72 Kimi K2.7 CodeMoonshot AI
67.4%
$0.95 / $4
74 Kimi K2.6Moonshot AI
65.9%
$0.95 / $4
75 GPT-5.5 (low reasoning)OpenAI
65.5%
$5 / $30
76 Qwen3.8-27B (medium reasoning)Alibaba
65.2%
$0.50 / $3
76 MiniMax-M3MiniMax
65.2%
$0.30 / $1.20
76 MiMo-V2.5-Pro (reasoning on)Xiaomi
65.2%
$0.43 / $0.87
79 DeepSeek-V4-Pro (0423, high reasoning)DeepSeek
64.8%
$1.42 / $2.83
80 Hy3Tencent
64.4%
$0.14 / $0.58
81 DeepSeek-V4-Pro (0423, max reasoning)DeepSeek
64.0%
$1.42 / $2.83
82 MiMo-V2.5Xiaomi
63.7%
$0.17 / $0.34
83 GPT-5.6 Terra (low reasoning)OpenAI
62.5%
$2 / $12
84 Muse SparkMeta
62.2%
–
85 DeepSeek-V4-Flash (0423, max reasoning)DeepSeek
61.8%
$0.14 / $0.28
85 GLM-5.1Z.ai
61.8%
$1.38 / $4.40
87 Qwen3.6-PlusAlibaba
61.4%
$0.33 / $1.95
88 Qwen3.7-PlusAlibaba
61.0%
$0.32 / $1.28
88 GPT-5.5 (no reasoning)OpenAI
61.0%
$5 / $30
90 Qwen3.6-27BAlibaba
60.7%
$0.30 / $3.20
90 GPT-5.4 nano (extra-high reasoning)OpenAI
60.7%
$0.20 / $1.25
92 GPT-5.4 mini (extra-high reasoning)OpenAI
59.2%
$0.75 / $4.50
93 Solar Pro 4Upstage
57.3%
$0.09 / $0.36
94 DeepSeek-V4-Flash (0423, high reasoning)DeepSeek
56.9%
$0.14 / $0.28
95 GPT-5.6 Terra (no reasoning)OpenAI
56.2%
$2 / $12
96 Claude Sonnet 4.5 (reasoning on)Anthropic
55.8%
$3 / $15
97 MiniMax-M2.7MiniMax
55.4%
$0.30 / $1.20
98 Inkling SmallThinking Machines
55.1%
$0.45 / $1.20
98 Inkling (extra-high reasoning)Thinking Machines
55.1%
$0.95 / $4.05
100 Gemini 3.5 Flash-LiteGoogle
53.6%
$0.30 / $2.50
101 GPT-5.6 Luna (medium reasoning)OpenAI
53.2%
$0.20 / $1.20
102 GPT-5.1 (high reasoning)OpenAI
52.4%
$1.25 / $10
103 Grok Build 0.1xAI
52.1%
$1 / $2
104 Muse Glimmer (high reasoning)Meta
51.7%
–
104 GLM-5.2 (no reasoning)Z.ai
51.7%
$1.40 / $4.40
106 Qwen3.5-397B-A17BAlibaba
51.3%
$0.55 / $3.50
106 Qwen3.6-27B (no reasoning)Alibaba
51.3%
$0.30 / $3.20
108 Mistral Medium 3.5Mistral AI
50.6%
$1.50 / $7.50
109 GLM-4.6 (reasoning on)Z.ai
49.4%
$0.50 / $2
110 Qwen3.8-27B (no reasoning)Alibaba
49.1%
$0.50 / $3
111 Qwen3.5-122B-A10BAlibaba
47.6%
$0.26 / $2.08
112 Qwen3.5-122B-A10B (no reasoning)Alibaba
47.2%
$0.26 / $2.08
113 DeepSeek-V3.2 (reasoning on)DeepSeek
46.8%
$0.30 / $0.96
114 Kimi K2.5Moonshot AI
45.7%
$0.57 / $2.85
115 GLM-4.7Z.ai
45.3%
$0.54 / $1.98
116 Qwen3.6-35B-A3BAlibaba
44.9%
$0.10 / $1
116 DeepSeek-V3.1-Terminus (reasoning on)DeepSeek
44.9%
$0.27 / $1
118 Claude Haiku 4.5 (reasoning on)Anthropic
44.2%
$1 / $5
119 Gemma 4 31BGoogle
43.4%
$0.14 / $0.40
119 GPT-5.6 Luna (low reasoning)OpenAI
43.4%
$0.20 / $1.20
121 Qwen3.6-35B-A3B (no reasoning)Alibaba
41.6%
$0.10 / $1
122 Qwen3.5-35B-A3B (no reasoning)Alibaba
40.8%
$0.16 / $1.30
123 Grok 4.3 (high reasoning)xAI
39.7%
$1.25 / $2.50
124 Gemma 4 26B A4BGoogle
39.0%
$0.10 / $0.30
124 GPT-5.6 Luna (no reasoning)OpenAI
39.0%
$0.20 / $1.20
126 Nemotron 3 Super 120B A12BNVIDIA
38.6%
$0.085 / $0.40
127 Qwen3-Coder-NextAlibaba
38.2%
$0.18 / $0.90
128 Claude Sonnet 4 (reasoning on)Anthropic
36.3%
$3 / $15
129 GPT-5 (high reasoning)OpenAI
35.2%
$1.25 / $10
130 GPT-5.5 Instant (2026-06-26)OpenAI
34.8%
–
131 Grok 4.3 (no reasoning)xAI
34.1%
$1.25 / $2.50
132 Gemini 3.1 Flash-Lite PreviewGoogle
31.1%
$0.25 / $1.50
133 Devstral 2Mistral AI
30.3%
$0.40 / $2
134 Devstral Small 2Mistral AI
29.6%
–
135 Qwen3.5-9BAlibaba
29.2%
$0.10 / $0.15
135 Gemma 4 31B (no reasoning)Google
29.2%
$0.14 / $0.40
137 Gemini 2.5 ProGoogle
28.5%
$1.25 / $10
138 Gemma 4 12BGoogle
27.3%
–
138 Mercury 2Inception
27.3%
$0.25 / $0.75
140 Mistral Small 3.1 24BMistral AI
26.2%
$0.35 / $0.56
140 gpt-oss-120b (high reasoning)OpenAI
26.2%
$0.15 / $0.60
142 Qwen3.5-4B (reasoning on)Alibaba
25.8%
–
143 Qwen3.5-4B (no reasoning)Alibaba
21.3%
–
143 Qwen3.5-9B (no reasoning)Alibaba
21.3%
$0.10 / $0.15
145 Mistral Small 4Mistral AI
21.0%
$0.15 / $0.60
146 Trinity Large ThinkingArcee AI
20.6%
$0.25 / $0.80
147 DeepSeek-R1DeepSeek
19.1%
$0.70 / $2.50
148 Granite 4.2 8BIBM
18.4%
$0.06 / $0.25
149 DeepSeek-V3DeepSeek
16.9%
$0.26 / $1.03
150 Nova 2 Lite (high reasoning)Amazon
16.1%
$0.30 / $2.50
151 DeepSeek-V3-0324DeepSeek
13.9%
$0.25 / $1
151 Mistral Medium 3.1Mistral AI
13.9%
$0.40 / $2
151 gpt-oss-120b (low reasoning)OpenAI
13.9%
$0.15 / $0.60
151 gpt-oss-20b (high reasoning)OpenAI
13.9%
$0.03 / $0.15
155 DiffusionGemma 26B A4BGoogle
12.4%
–
155 Magistral Medium 1.2Mistral AI
12.4%
–
157 Qwen3-235B-A22B-Thinking-2507Alibaba
12.0%
$0.30 / $3
157 Mistral Large 3Mistral AI
12.0%
$0.50 / $1.50
157 Solar Pro 3Upstage
12.0%
$0.15 / $0.60
160 GPT-4.1 miniOpenAI
10.1%
$0.40 / $1.60
161 Ministral 3 14BMistral AI
9.7%
$0.20 / $0.20
162 Llama 4 MaverickMeta
7.9%
$0.27 / $0.85
163 Qwen3-Next-80B-A3B-ThinkingAlibaba
6.7%
$0.15 / $1.20
163 Nemotron 3 Nano 30B A3B (reasoning on)NVIDIA
6.7%
$0.05 / $0.20
165 Mistral Small 3.2 24BMistral AI
5.6%
$0.094 / $0.25
165 GPT-4o mini (2024-07-18)OpenAI
5.6%
$0.15 / $0.60
167 Qwen3-32B (reasoning on)Alibaba
5.2%
$0.14 / $0.40
168 Qwen3-14B (reasoning on)Alibaba
4.9%
$0.12 / $0.24
168 Llama 3.3 70B InstructMeta
4.9%
$0.59 / $0.79
170 Gemma 3 27BGoogle
4.5%
$0.12 / $0.20
170 Magistral Small 1.2Mistral AI
4.5%
–
170 o3-mini (high reasoning)OpenAI
4.5%
$1.10 / $4.40
173 Ministral 3 8BMistral AI
4.1%
$0.15 / $0.15
174 Llama 4 ScoutMeta
3.7%
$0.18 / $0.59
174 GPT-4.1 nanoOpenAI
3.7%
$0.10 / $0.40
174 GPT-5 mini (high reasoning)OpenAI
3.7%
$0.25 / $2
177 Qwen3.5-2B (reasoning on)Alibaba
3.0%
–
178 Qwen3-8B (reasoning on)Alibaba
2.2%
$0.12 / $0.46
179 Gemma 4 E4BGoogle
1.9%
–
180 Qwen3-30B-A3B-Thinking-2507Alibaba
1.5%
$0.20 / $2.40
180 Llama 3.1 8B InstructMeta
1.5%
$0.05 / $0.08
182 Qwen3.5-0.8B (no reasoning)Alibaba
0.4%
–
182 Gemma 3 4BGoogle
0.4%
$0.05 / $0.10
182 Gemma 4 E2BGoogle
0.4%
–
185 Qwen3.5-0.8B (reasoning on)Alibaba
0.0%
–
185 Qwen3.5-2B (no reasoning)Alibaba
0.0%
–
185 Gemma 3 12BGoogle
0.0%
$0.05 / $0.15
185 Ministral 3 3BMistral AI
0.0%
$0.10 / $0.10

Swipe the table sideways for more columns.

Ranks follow the score as shown, so equal numbers share a rank. Results as published by Artificial Analysis; we do not re-run them.

Progress over time

Terminal-Bench 2.1, plotted at each model's release date, higher is better. Each model counts once, at its best setting. Scores are today's, not launch day's. Hover or tap a mark for its model.

Best score so far

Every model at its release date. The line moves each time a new release beats every earlier one, and each new best is named where there is room.

-50%0%50%100% 2025Jul2026Jul GPT-4o mini (2024-07-18)DeepSeek-V3DeepSeek-R1Mistral Small 3.1 24BGemini 2.5 ProClaude Sonnet 4DeepSeek-V3.1-TerminusClaude Sonnet 4.5Gemini 3.1 Pro PreviewGPT-5.4GPT-5.5Claude Opus 4.8GPT-5.6 SolClaude Fable 5.1

Best score by lab

Each lab's best so far, stepping whenever one of its releases beats its own earlier models. The 6 labs with the best scores today; 14 others not drawn.

AnthropicOpenAIAlibabaxAIGoogleMeta

0%50%100% 2025Jul2026Jul AnthropicOpenAIAlibabaxAIGoogleMeta

What it measures

Share of Terminal-Bench 2.1 tasks completed, as run by Artificial Analysis with its own harness and prompts.

What it does not measure

Not other harnesses or tools; superseded by 4.0 for newer models.

Source

Artificial Analysis · Methodology

285 results from Artificial Analysis not ranked here · show why

We rank a result only when we can tie it to a specific model you can use. These are left out:

  • Not on sale through the API providers we track: 270
  • A different snapshot or variant from the model we list: 14
  • An unusual combination of settings: 1

Data sourced from Artificial Analysis. Licence: Artificial Analysis commercial data licence.