Best Value LLM for Coding in 2026 — Cost-Adjusted Rankings

Raw benchmark scores only tell half the story. This ranking divides each model's weighted coding score by its output token price, surfacing models that deliver the most coding capability per dollar spent. A model scoring 70 at $1/1M tokens outranks one scoring 80 at $15/1M tokens here — because for the same budget you get far more coding work done. Use this alongside the standard coding leaderboard to find the sweet spot between performance and cost for your coding workflows.

According to BenchLM.ai, Gemini 3.1 Flash-Lite leads this ranking with a score of 76.62, followed by DeepSeek Coder 2.0 (47.76) and Gemini 2.5 Flash (39.01). There is a significant gap between the leading models and the rest of the field.

The best open-weight option is DeepSeek Coder 2.0 (ranked #2 with a score of 47.76). Open-weight models are highly competitive in this category — self-hosting is a viable alternative to proprietary APIs.

This ranking is based on weighted averages across the scoring benchmarks in coding tracked by BenchLM.ai. For detailed model profiles, click any model name below. To compare two specific models head-to-head, use the "vs #" links.

Gemini 3.1 Flash-Lite
GoogleProprietary1M

76.62

Score/$

Score: 30.6 · $0.4/1M

DeepSeek Coder 2.0
DeepSeekOpen Weight128K

47.76

Score/$

Score: 52.5 · $1.1/1M

Gemini 2.5 Flash
GoogleProprietary1M

39.01

Score/$

Score: 23.4 · $0.6/1M

4
DeepSeek V3
DeepSeekOpen Weight128K

30.81

Score/$

Score: 33.9 · $1.1/1M

5
Kimi K2.5
Moonshot AIOpen Weight128K

24.82

Score/$

Score: 69.5 · $2.8/1M

6
Gemini 3 Flash
GoogleProprietary1M

18.1

Score/$

Score: 54.3 · $3/1M

7
GPT-4.1 mini
OpenAIProprietary1M

17.74

Score/$

Score: 28.4 · $1.6/1M

8
Gemini 3.1 Pro
GoogleProprietary1M

15.56

Score/$

Score: 77.8 · $5/1M

9
DeepSeek-R1
DeepSeekOpen Weight128K

14.47

Score/$

Score: 31.7 · $2.19/1M

10
Claude Haiku 4.5
AnthropicProprietary200K

12.11

Score/$

Score: 48.5 · $4/1M

11
GPT-5.1
OpenAIProprietary200K

11.75

Score/$

Score: 70.5 · $6/1M

12
o3-mini
OpenAIProprietary200K

10.37

Score/$

Score: 45.6 · $4.4/1M

13
Gemini 2.5 Pro
GoogleProprietary1M

9.71

Score/$

Score: 48.5 · $5/1M

14
GPT-5.2
OpenAIProprietary400K

9.45

Score/$

Score: 75.6 · $8/1M

15
GPT-5.1-Codex-Max
OpenAIProprietary400K

9.28

Score/$

Score: 74.2 · $8/1M

16
GPT-5.2-Codex
OpenAIProprietary400K

9.15

Score/$

Score: 73.2 · $8/1M

17
GPT-5.3 Codex
OpenAIProprietary400K

7.51

Score/$

Score: 75.1 · $10/1M

18
Mistral Large 3
MistralProprietary128K

7.05

Score/$

Score: 42.3 · $6/1M

19
GPT-4.1
OpenAIProprietary1M

6.14

Score/$

Score: 49.1 · $8/1M

20
GPT-5.4
OpenAIProprietary1.05M

5.07

Score/$

Score: 76.1 · $15/1M

21
Claude Sonnet 4.6
AnthropicProprietary200K

4.95

Score/$

Score: 74.2 · $15/1M

22
Grok 4.1
xAIProprietary1M

4.93

Score/$

Score: 73.9 · $15/1M

23
Claude Sonnet 4.5
AnthropicProprietary200K

4.06

Score/$

Score: 60.8 · $15/1M

24
GPT-4o
OpenAIProprietary128K

2.99

Score/$

Score: 29.9 · $10/1M

25
o3
OpenAIProprietary200K

1.47

Score/$

Score: 59 · $40/1M

26
Claude Opus 4.6
AnthropicProprietary1M

1.06

Score/$

Score: 79.3 · $75/1M

27
o1
OpenAIProprietary200K

0.74

Score/$

Score: 44.2 · $60/1M

28
GPT-5.4 Pro
OpenAIProprietary1.05M

0.49

Score/$

Score: 88.3 · $180/1M

Key Takeaways

  • The best value model is Gemini 3.1 Flash-Lite by Google with a Score/$ ratio of 76.62 (score: 30.6, output: $0.4/1M tokens).
  • The best open-weight model in this ranking is DeepSeek Coder 2.0 at position #2.
  • 28 models are included in this ranking.
Last updated: April 3, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.