- Rank
- #159
- Total Output Tokens
- 69,489
- Response Time (avg)
- 17.13s
- Total Cost
- $3.787
Claude Opus 4.8 (low) vs DeepSeek V4 Flash 0423 (high)
DeepSeek V4 Flash 0423 (high) leads on average score with 8.1 vs 7.2. DeepSeek V4 Flash 0423 (high) has the lower benchmark cost at $0.261 vs $3.787. Claude Opus 4.8 (low) is faster at 17.13s vs 55.99s, with pass rates of 78.3% vs 71.0%.
Compared models
- Rank
- #97
- Total Output Tokens
- 203,134
- Response Time (avg)
- 55.99s
- Total Cost
- $0.261
Recommended model
DeepSeek V4 Flash 0423 (high)
It has the best score here (8.1), while costing about 14.5x less than Claude Opus 4.8 (low).
Detailed comparison
| Metric | Claude Opus 4.8 Claude Opus 4.8 low | DeepSeek V4 Flash 0423 DeepSeek V4 Flash 0423 high |
|---|---|---|
| Score | 7.2 | 8.1 |
| Rank | #159 | #97 |
| Reliability | 10.0 | 10.0 |
| Consistency | 8.6 | 8.6 |
| Attempts | 69/69 | 69/69 |
| Tests Correct | ||
| Attempt pass rate | 78.3% | 71.0% |
| Flaky tests | 4 | 4 |
| Total Runs | 69 | 69 |
| Cost per result | 23.669 | 0.547 |
| Total Cost | $3.787 | $0.261 |
| Input Price | $5.000 / 1M | $0.002 / 1M |
| Output Price | $25.000 / 1M | $1.280 / 1M |
| Cache Read Price | $0.500 / 1M | $0.002 / 1M |
| Cache Write Price | $6.250 / 1M | N/A |
| Total Input Tokens | 409,947 | 298,794 |
| Output Tokens | 46,060 | 45,873 |
| Reasoning Tokens | 23,429 | 188,721 |
| Response Time (avg) | 17.13s | 55.99s |
| Response Time (max) | 127.97s | 218.13s |
| Response Time (total) | 394.01s | 1287.87s |
| Parameters | ~5T total (~500B active) | 284B total (13B active) |
| Availability | Closed | Open source |
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#159 Claude Opus 4.8
low- Cost
- $0.031
- Time
- 14.1s
- Tokens
- 1,345 tok
#97 DeepSeek V4 Flash 0423
high- Cost
- $0.003
- Time
- 93.1s
- Tokens
- 7,926 tok
Category Breakdown
| Agentic | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 4.7 | 3.1 | 33.3% | 1 | 110.68s | 253,425 | 3,004 | 14,248 | |
| DeepSeek V4 Flash 0423 | 10.0 | 10.0 | 100.0% | 0 | 148.13s | 190,314 | 8,644 | 14,640 |
| Anti-AI Tricks | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 10.0 | 10.0 | 100.0% | 0 | 3.30s | 834 | 793 | 371 | |
| DeepSeek V4 Flash 0423 | 8.3 | 10.0 | 75.0% | 0 | 28.51s | 540 | 140 | 7,770 |
| Coding | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 6.6 | 4.6 | 77.8% | 2 | 7.58s | 10,590 | 3,637 | 809 | |
| DeepSeek V4 Flash 0423 | 7.8 | 10.0 | 66.7% | 0 | 50.60s | 7,279 | 395 | 34,862 |
| Combined | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 9.9 | 10.0 | 100.0% | 0 | 36.87s | 119,079 | 13,586 | 2,867 | |
| DeepSeek V4 Flash 0423 | 6.4 | 5.8 | 66.7% | 1 | 104.10s | 82,663 | 4,633 | 37,533 |
| Data parsing and extraction | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 6.3 | 5.8 | 66.7% | 1 | 2.27s | 10,503 | 310 | 0 | |
| DeepSeek V4 Flash 0423 | 10.0 | 10.0 | 100.0% | 0 | 28.03s | 7,290 | 201 | 1,179 |
| Domain specific | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 5.3 | 10.0 | 33.3% | 0 | 46.54s | 972 | 23,226 | 4,472 | |
| DeepSeek V4 Flash 0423 | 3.6 | 7.2 | 22.2% | 1 | 115.42s | 754 | 22,778 | 79,643 |
| General Intelligence | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 10.0 | 10.0 | 100.0% | 0 | 2.55s | 708 | 231 | 0 | |
| DeepSeek V4 Flash 0423 | 6.1 | 3.1 | 66.7% | 1 | 25.15s | 471 | 79 | 632 |
| Instructions following | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 9.8 | 10.0 | 100.0% | 0 | 2.78s | 909 | 111 | 221 | |
| DeepSeek V4 Flash 0423 | 10.0 | 10.0 | 100.0% | 0 | 15.36s | 627 | 63 | 1,622 |
| Puzzle Solving | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 10.0 | 10.0 | 100.0% | 0 | 3.01s | 894 | 592 | 184 | |
| DeepSeek V4 Flash 0423 | 8.2 | 7.2 | 88.9% | 1 | 26.11s | 594 | 196 | 1,767 |
| Tool Calling | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 10.0 | 10.0 | 100.0% | 0 | 6.85s | 11,775 | 370 | 35 | |
| DeepSeek V4 Flash 0423 | 10.0 | 10.0 | 100.0% | 0 | 74.73s | 8,079 | 228 | 542 |
| Trivia | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 | 3.0 | 10.0 | 0.0% | 0 | 5.48s | 258 | 200 | 222 | |
| DeepSeek V4 Flash 0423 | 3.0 | 10.0 | 0.0% | 0 | 54.46s | 183 | 8,516 | 8,531 |
Quick Compare
Switch Comparison Pair
Claude Opus 4.7mediumvsDeepSeek V4 Flash 0423highClaude Opus 4.8lowvsQwen3.8 27BmediumFree AvailableClaude Opus 4.8lowvsQwen3.7 FlashhighClaude Opus 4.8lowvsQwen3.5 Plus 2026-02-15mediumClaude Opus 4.8lowvsLongCat 2.0mediumClaude Opus 4.8lowvsGLM 5mediumClaude Opus 4.8lowvsQwen3.5-27BnoneClaude Opus 4.8lowvsQwen3.6 27BmediumSeed-2.0-CodelowvsDeepSeek V4 Flash 0423highClaude Opus 4.8lowvsGemini 3.5 Flash LitemediumDeepSeek V4 Flash 0423highvsNemotron 3 UltramediumFree AvailableClaude Opus 4.8lowvsQwen3.6 Plusmedium